TYPE Original Research PUBLISHED 27 February 2023 DOI 10.3389/fpsyg.2023.1127193 The Conners Continuous OPEN ACCESS Performance Test CPT3™: Is it a reliable marker to predict EDITED BY Fabrizio Stasolla, Giustino Fortunato University, Italy REVIEWED BY neurocognitive dysfunction in Khalida Akbar, Durban University of Technology, South Africa Myalgic encephalomyelitis/ Maurel Sara, Autonomous University of Barcelona, Spain chronic fatigue syndrome? *CORRESPONDENCE Judith Fernández-Quirós Judith Fernández-Quirós 1,2*, Marcos Lacasa-Cazcarra 3, judith.fquiros@vallhebron.cat Jose Alegre-Martín 1, Ramón Sanmartín-Sentañes 1, SPECIALTY SECTION Miriam Almirall 1, Patricia Launois-Obregón 1, This article was submitted to Neuropsychology, Jesús Castro-Marrero 1, Amanda Rodríguez-Urrutia 2, 4,5,6, a section of the journal Frontiers in Psychology Jose A. Navarro-Sanchis 2,4 and J. Antoni Ramos-Quiroga 2,4,5,6 RECEIVED 19December 2022 1 Myalgic Encephalomyelitis/Chronic Fatigue Syndrome Unit, Division of Rheumatology, Vall d’Hebron ACCEPTED 07 February 2023 Hospital Research Institute Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Department of PUBLISHED 27 February 2023 Mental Health, Hospital Universitari Vall d’Hebron, Barcelona, Spain, 3 e-Health Center, Universitat Oberta de Catalunya, Barcelona, Spain, 4 Group of Psychiatry, Mental Health and Addictions, Vall CITATION d’Hebron Research Institute (VHIR), Barcelona, Spain, 5 Biomedical Network Research Centre on Mental Fernández-Quirós J, Lacasa-Cazcarra M, Health (CIBERSAM), Barcelona, Spain, 6 Department of Psychiatry and Forensic Medicine, Universitat Alegre-Martín J, Sanmartín-Sentañes R, Autònoma de Barcelona, Barcelona, Spain Almirall M, Launois-Obregón P, Castro-Marrero J, Rodríguez-Urrutia A, Navarro-Sanchis JA and Ramos-Quiroga JA (2023) The Conners Continuous Performance Test CPT3™: Is it a Introduction: The main objective is to delimit the cognitive dysfunction reliable marker to predict neurocognitive associated with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/ dysfunction in Myalgic encephalomyelitis/ chronic fatigue syndrome? CFS) in adult patients by applying the Continuous Performance Test (CPT3™). Front. Psychol. 14:1127193. Additionally, provide empirical evidence on the usefulness of this computerized doi: 10.3389/fpsyg.2023.1127193 neuropsychological test to assess ME/CFS. COPYRIGHT © 2023 Fernández-Quirós, Lacasa-Cazcarra, Method: The final sample (n =225; 158 Patients/67 Healthy controls) were Alegre-Martín, Sanmartín-Sentañes, Almirall, recruited in a Central Sensitization Syndromes (CSS) specialized unit in a tertiary Launois-Obregón, Castro-Marrero, Rodríguez- hospital. All participants were administered this neuropsychological test. Urrutia, Navarro-Sanchis and Ramos-Quiroga. This is an open-access article distributed under Results: There were significant differences between ME/CFS and healthy controls the terms of the Creative Commons Attribution in all the main measures of CPT3™. Mainly, patients had a worse indicator of License (CC BY). The use, distribution or reproduction in other forums is permitted, inattentiveness, sustained attention, vigilance, impulsivity, slow reaction time, and provided the original author(s) and the more atypical T-scores, which is associated with a likelihood of having a disorder copyright owner(s) are credited and that the characterized by attention deficits, such as Attention Deficit Hyperactivity original publication in this journal is cited, in accordance with accepted academic practice. Disorder (ADHD). In addition, relevant correlations were obtained between the No use, distribution or reproduction is CPT3™ variables in the patient’s group. The most discriminative indicators of ME/ permitted which does not comply with these CFS patients were Variability and Hit Reaction Time, both measures of response terms. speed. Conclusion: The CPT3™ is a helpful tool to discriminate neurocognitive impairments from attention and response speed in ME/CFS patients, and it could be used as a marker of ME/CFS severity for diagnosing or monitoring this disease. KEYWORDS neuropsychological test, cognitive impairments, continuous performance test, CPT3™, neurocognitive dysfunction, Central Sensitization Syndromes, chronic fatigue syndrome Frontiers in Psychology 01 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 Introduction cognitive difficulties, and those with greater pain were associated with lower objective working memory function. ME/CFS patients primarily Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) had psychomotor speed and attention problems, measured by is a complex multisystem disease that presents a chronic course with objective neuropsychological tests (Rasouli et al., 2019). Other studies periods of symptomatic exacerbation frequently related to acute stress. confirm that cognitive deficits in ME/CFS depend mainly on This condition predominantly affects women and is a severe functional compromised attention, memory, and reaction time, but motor disorder (Afari and Buchwald, 2003; Prins et al., 2006). The prevalence functioning, vocabulary, reasoning, and global deficits are absent is estimated to be between 0.2 and 2.6% of the general population (Cockshell and Mathias, 2010). (Reyes et al., 2003; Nacul et al., 2011). The nuclear symptoms are Cognitive dysfunction has been described in patients with chronic central fatigue (>6 months) and post-exertional malaise with depression and anxiety, and it is difficult to determine whether this is a recovery time longer than 24 h of idiopathic origin. The international due to psychopathological comorbidity or fatigue. Slowing diagnostic criteria were established by the Centres for Disease Control information processing speed occurs, especially in complex tasks (CDC) in Atlanta (Georgia) in 1994 (Fukuda et al., 1994). requiring sustained attention. However, the underlying mechanisms Furthermore, the heterogeneity of symptoms inME/CFS is studied in of the manifested cognitive dysfunction remain unclear (Cvejic et al., clusters from 2003 as a complement to the CDC’s criteria 2016), advising a need for appropriate neuropsychological assessment (neurological, muscle, cognitive, neurovegetative, and immunological; tools in this complex disease capable of overcoming such limitations. Fukuda et al., 1994; Carruthers et al., 2003). In 2011, these criteria One study showed that the dysfunction of information processing were updated, and post-exertional exhaustion was proposed as a speed is independent of depressive symptoms in ME/CFS, suggesting disease hallmark (Carruthers et al., 2011). ME/CFS is associated with that attentional deficits may be primary to memory problems, which different comorbid phenomena (anxiety-depressive disorders, could show an underlying neurological basis for the attentional fibromyalgia (FM), sicca syndrome, regional myofascial pain dysfunction in these patients (Santamarina-Perez et al., 2014). syndrome, plantar fasciitis, degenerative or mechanical disk disease, Unfortunately, there are currently no commercially available and tendinopathy of the shoulder; Ruiz et al., 2011 Castro-Marrero diagnostic tests, specific lab biomarkers, or targeted FDA-approved et al., 2016) that are more prevalent in ME/CFS patients than in ME/CFS drugs (Castro-Marrero et al., 2017). For all these arguments, non-CFS individuals (Rivera et al., 2006). At present, the most widely and given the disparity of the published studies, there is no consensus accepted hypothesis for the pathogenesis of ME/CFS characterizes it on the results regarding cognitive dysfunction in ME/CFS and its as a genetic-based process with different triggering factors and repercussions. Therefore, more research is needed on this symptom in subsequent neuroimmunology and immunoinflammatory ME/CFS and the detection of the underlying mechanisms. The dysfunction (Bassi et al., 2008). limitations found in previous studies on this topic must be considered Fatigue is the most common symptom associated with chronic to improve their research and obtain more specific neuropsychological diseases and is experienced as really distressing. Nature fatigue is a assessment tools to evaluate it (Beaumont et al., 2012). General subjective state with both physical and psychological elements, and dysfunction in attention and information processing speed have been there is a lack of effective treatments for it. New methods are being suggested as reasons for ME/CFS memory complaints (Deluca et al., developed to quantify fatigue and are increasing in clinical settings 2004; Dickson et al., 2009). (Swain, 2000). Fatigue may occur due to a physical or psychological Considering the evidence of neuropsychological dysfunction in event, or fatigue may cause a physical event. The concept of fatigue ME/CFS patients, this work is carried out using a computerized test appears to be defined by overlapping terms of cognitive or mental that allows comprehensive measurement of different cognitive fatigue; however, the description for each slightly varies. For example, domains of interest in this disease. Therefore, this is the first study in cognitive fatigue is said to occur when cognitive performance ME/CFS patients using the CPT3™. This neuropsychological test has decreases by engaging in tasks requiring sustained activity (Morrow been widely used to objectively and comprehensively assess cognitive et al., 2015). Mental fatigue has been defined as a subjective feeling of functioning in patients diagnosed with Attention-Deficit Hyperactivity tiredness and inertia that occurs during extended periods of Disorder (ADHD; Fasmer et al., 2016; Berger et al., 2017). demanding cognitive activity (Badin et al., 2016). Fatigue has also CPT3™ allows evaluating the participant’s performance in just been defined as physiological fatigue, which is described as muscle 14 min, contemplating the “fatigue” variable (the core symptom of this weakness that may occur due to exercise (Prinsen et al., 2015). There disease) and other multiple cognitive measures. It is not only is insufficient evidence examining the relationship between fatigue interesting because of the novelty of introducing this software tool in and cognitive impairments in patients with chronic illnesses. More patients with ME/CFS, but also because it allows the evaluation of studies are needed to examine the potential relationships between multiple indicators of cognitive functioning such as attentional these two symptoms to develop effective treatments for individuals capacity and some executive functions (inhibition and information living with NCDs (Chronic Non-Communicable Diseases; Menzies processing speed) trying to find a differential clinical profile between et al., 2021). patients and healthy controls (Kao and Thomas, 2010). Cognitive deficits are common in ME/CFS patients and limit their quality of life and psychological well-being. The discrepancy between self-reported common cognitive deficits and results obtained by Method objective neuropsychological tests has been reported (Rasouli et al., 2019). This disparity could be explained by several reasons, mainly Sample methodological (Cockshell and Mathias, 2010), suggesting establishing an appropriate method to avoid it. Patients with higher The study subjects were composed of two groups, healthy controls levels of fatigue, pain, and depression reported greater subjective and patients diagnosed with ME/CFS. Patients with a diagnosis of Frontiers in Psychology 02 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 ME/CFS have been recruited consecutively from the Central was performed from 9:00 am to 2:00 pm, always in the same medical Sensitization Syndromes (SCC) specialized unit at the Vall d’Hebron office under standard temperature, noise, and lighting conditions, and University Hospital in Barcelona from July 2021 to March 2022. with mobile turned off to avoid interference. It was decided to establish Inclusion criteria were fulfilled the Fukuda et al. (1994) and Carruthers this time slot for the evaluation due to several reasons (for example, criteria for ME/CFS(2), age between 18 and 65 years, the subjective attentional, and cognitive performance is not as optimal at certain clinical manifestation of cognitive impairment according to times of the day, and fatigue accumulates as the day progresses both Carruthers et al. (2003, 2011), understanding and acceptance of in the clinical population and in healthy controls as a general rule). participation in the study and signed written informed consent. The The evaluation protocol was not modified for any patient. research protocol was approved by the ethics committee with this CPT3™ is a computerized task of continuous execution based on registration code(PR(AG)257/2021). The patients underwent an performance to assess primarily attention-related impairments in extensive medical examination by specialist physicians belonging to individuals aged 8 years and older. The administration time is 14 min, the Central Sensitization Syndromes (CSS) experienced Unit of the during which the subject must maintain attention to perform this task. Rheumatologist Department at this hospital. The exclusion criteria There are six blocks of trials, with three sub-blocks, each consisting of were difficulty in understanding and/or completing the self-reported 20 trials. Each stimulus appears on the screen with varying time questionnaires, the presence of severe unstable psychiatric disorders frequencies (inter-stimulus intervals from 1, 2, and 4 s), which allows (such as psychotic episodes, major depressive episodes, manic for comparison of the subject’s attention and response speed according episodes, and anorexia nervosa), history of neurological disorders to the time intervals. This software makes it possible to compare the with cognitive impairment (such as severe head-brain trauma), changes in performance that may be experienced in separate segments presence of another systemic disease with cognitive alterations and (block measurement), thus checking whether the level of vigilance presence of the disease entity Mild Cognitive Impairment (MCI). All fluctuates during this simple task. Its duration allows for controlling patients diagnosed with ME/CFS included in the study were recorded the “fatigue effect” on cognitive performance, a key aspect to consider on a data collection sheet in encrypted form and subsequently in patients diagnosed with CFS. Before starting, some instructions included in a database. The results obtained from the sample through must be given. Individuals are seated in front of a computer. Each the CPT3™ software were recorded in an excel table. The minimum participant is required to respond (pressing the space bar on the sample calculation to demonstrate a statistically significant difference keyboard) when any letter (target stimuli), except the letter “X” between the mean of the measurements of the cognitive domain (non-target stimuli), appears on the screen. It is essential to warn them variable, information processing speed (HRT), according to Student’s that they must keep responding until the end of the test to obtain the t-test, is 92 patients with a confidence level equal to 0.95. Pearson’s computer-generated reports describing the respondent’s performance coefficient, which measures the variance to the mean, is 20%. in detail. Once the subjects have understood the instructions, they Of the total initial sample, four patients were excluded because of perform the test. Each participant’s final report was recorded in an an associated severe psychiatric or neurological comorbidity at that Excel spreadsheet. Subsequently, the neurocognitive profile of the precise moment that could account for some of the ME/CFS whole sample was studied by analyzing all the results obtained (Kao symptoms (three had a major depressive episode, and one patient with and Thomas, 2010). severe head-brain trauma). In addition, six patients were also excluded because they withdrew on their own initiative, indicating fear of fatigue from the mental overexertion of the test. Main features of the CPT3™ The CPT3™ test offers already standardized and raw scores to Neuropsychological evaluation determine not only the overall performance of the evaluation but also specify different types of attention deficits (e.g., inattentiveness, In the first session, the participants were individually examined impulsivity, sustained attention, vigilance) and response speed, by the ME/CFS specialist physician and were given questionnaires to allowing a comprehensive assessment. So, multiple features are fill in. The perception of disabling fatigue, sleep problems, and health- measured in The CPT3™. related quality of life using self-administered questionnaires: the A brief description of the main measures used in this study is fatigue impact scale FIS40 (Fisk et al., 1994), FIS8 (Fisk and Doble, exposed below: 2002), PSQI (Verster et al., 2008), the Short Form Health Survey (SF-36; Alonso et al., 1995), the Symptom Checklist-90-revised (SCL • Response speed: to obtain information performance about motor 90 R) psychological inventory (McGregor et al., 1997) and hospital reaction time and information processing speed measured by the anxiety and depression scale (HAD; Castresana et al., 1995). The software (HRT, HRTSD, HRT block change, …). interviews were conducted by two internists and one rheumatologist • Impulsivity: is an indicator of the response inhibition capacity of physician skilled in diagnosing and treating this syndrome. the evaluated and includes a faster than normal HRT and a The neuropsychological evaluation was performed in a second higher than average rate of commissions and/or perseverations. appointment. It consisted of a single standardized individualized • Inattentiveness: scores about focused attention. This indicator examination in which cognitive functioning was assessed using the relates to poor detectability, a high percentage of omissions and latest version of the 3rd Conners Continuous Performance Test commissions, a slow HRT, and high levels of inconsistency in CPT3™ (Kao and Thomas, 2010). The participants had never been response speed. administered the CPT3™ to avoid familiarity. They were required to • Sustained attention: is defined as the respondent ability to respond when any letter except “X” appeared on the monitor. CPT3™ maintain attention as the administration progresses. A decrease Frontiers in Psychology 03 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 in sustained attention across time is captured by atypical slowing (control group of healthy patients) and group P (diagnosed group of in the respondent’s HRT and by increases in omissions and ME/CFS patients). The significance threshold has been defined as a commissions in later blocks of the administration. p-value <0.05. Cramer’s V is an effect size measurement for the • Vigilance: relates to the respondent’s performance at varying chi-square test of independence. It measures how strongly two levels of stimulus frequency and is defined by the respondent’s categorical fields are associated. The degree of freedom (df) is 1, and ability to maintain a performance level even when the task rate is if Cramer’s value >0.1 is considered small-medium, >0.30 is slow. It is captured by changes in the respondent’s HRT, as considered medium-large, and > 0.50 is considered significant (Kim, indicated by the variables HRT ISI. 2017). Cramer’s V was based on Pearson’s chi-squared statistic and • Detectability: discrimination between non-targets and targets. was published by Harald Cramer in 1946 (Doob, 1946). • Omissions: are missed targets and are generally an indicator of inattentiveness. Numerical features • Commissions: are incorrect responses to non-targets. High The Shapiro–Wilk test and D′Agostino’s K-squared test are commission error rates may indicate either inattentiveness or two of the most commonly used hypothesis tests to analyze impulsivity, depending on the respondent’s HRT. normality. In both tests, the null hypothesis is that the data comes • Perseverations are responses made in less than 100 ms following from a normal distribution. The p-value of these tests indicates the presentation of a stimulus. Perseverations may be related to the probability of obtaining data like the observed data if they impulsivity or an extremely liberal response style. came from a population with a normal distribution with the same • Hit reaction time (HRT): the mean response speed, measured in mean and variance as the observed data. The threshold is a p-value milliseconds, for all non-preservative responses made during the <0.05 as sufficient evidence to reject normality. The purpose is to entire administration. An atypically slow HRT may indicate verify the conditions of parametric methods for using the t-test. inattentiveness but may also result from a conservative response For features with a non-normal distribution, the test of style. So, HRT is also affected by response style. independence used is the Mann–Whitney U-test for independent • HRT standard deviation (HRT SD): measures the consistency of samples. The analysis is continued by calculating the arithmetic response speed for the entire administration. A high HRT SD means of each variable and the graphical calculation using the box indicates a greater response speed inconsistency, sometimes plot analysis. indicative of inattentiveness. • Variability: is a measure of response speed consistency; however, Relationship graph between features Variability is a “within respondent” measure (i.e., the amount of The relationships between the features have been analyzed using variability the respondent showed in 18 separate sub-blocks of graph theory. A graph is a collection of nodes (also called vertices) the administration with the overall HRT SD score). High joined together in pairs by edges (undirected) or arcs (directed; response speed variability indicates that the respondent’s Newman, 2018). The graph structure allows us to capture the pattern attention and information processing efficiency varied of interactions between the nodes (individuals or entities). Graph (or throughout the administration. network) analysis is used to study relationships between individuals • HRT block change: the slope of change in HRT across the six blocks to discover knowledge about global and local structures. The study of of the administration. A positive slope indicates decelerating HRT structure networks helps to decide the optimal order (Hagberg et al., as the administration progressed; a negative slope indicates 2008). In this work, the graph nodes are defined as all features, and the accelerating HRT; a flat slope indicates no change in HRT. edges are defined as moderate or strong correlations between nodes • HRT Inter-Stimulus Interval (ISI) change: is the slope of change (features). The correlation between two features is represented by in reaction time across the three ISIs (1, 2, and 4 s). A positive corr ( i,j ) , and Spearman correlation is defined as moderate or strong slope indicates decelerating HRT at longer intervals, whereas a if corr ( i,j ) ≥ 0.5 (Suchowski, n.d.) in case of direct correlation. It has negative slope indicates accelerating HRT at longer intervals. been created an edge ( i,j ) if abs ( corr ( i,j ) ) ≥ 0.5 in order to include direct and indirect correlation. This analysis used the values of the Based on the respondent’s pattern and scores in each attentional P-group diagnosed by ME/CFS. dimension, the software identifies the presence and severity of the kinds of attention problems the respondent is most likely having. Random forest algorithm Therefore, CPT3™ provides in the outcome report the likelihood of Random forest is a widespread algorithm for classification and having a disorder characterized by attention deficits, such as ADHD offers the importance of feature values. All classified samples from the (Kao and Thomas, 2010). two defined groups are used to obtain a model. This model classifies In this research, a categorical dichotomous variable (YES/NO; future samples according to both categorical and numerical CPT3 obtained from the assessment report) is analyzed to indicate whether scores. The metric used was accuracy. Features importance are those evaluated have this probability of having an attentional disorder computed as the mean and standard deviation of accumulation of the (YES indicates moderate to high possibility). impurity decrease within each tree. It informs which feature has greater power to classify (Liaw and Wiener, n.d.). Statistical method Results Categorical features For each feature, a 2×2 contingency table is set up to perform the The sample size’s value was based on a t-student analysis for two chi-squared test of independence for the groups defined as group C populations, ME/CFS patients (P) and healthy controls (C). Frontiers in Psychology 04 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 Distributions of CPT3™ variables were examined before analysis. in classification and regression problems. This model classifies if Sample Analysis is reported in Table 1. a patient is in a control group or not. The database is divided into There are 17 features analyzed, six categorical and 11 numerical. 75% to train the model and provide a generalization to predict, For categorical features, the Chi-Squared test of independence is used based on the values of the continuous and categorical variables, related to C (Control group) or P (Diagnosed group). Tests of whether it belongs to group P or group C. The remaining 25% is independence (Chi-Square, T-Student, and Mann–Whitney) indicate kept for testing the model, i.e., the model is run and compared to if there are significant differences between the two populations of the see how well the model fits the sample. The model offers 77.19% study. The results are shown in Tables 2, 3. accuracy, which means the 25% test dataset predicted with As a result, a pattern of attention deficit is obtained in the ME/CFS 77.19% accuracy. After dependence analysis for continuous group when recording as a categorical dichotomous variable (YES/ No) the probability of having a disorder characterized by attention TABLE 2 Results from numerical features from CPT3™ by groups. deficits, such as ADHD (YES, including moderate to high likelihood of having it). With T-Student (if normality), the detectability feature Feature Control Diagnosed Independent is significant. With Mann–Whitney U Test (not normality), all group (C) group (P) test variables are significant except HRT Blo C and HRT ISI Ch N, which Mean STD Mean STD p-Value are not. t-student test (for normal distribution variables) The categorical variables are dichotomous, with values {1,2}, the  Res.Styl N 47.82 8.05 48.95 11.21 0.39644 plots showing whether significant differences exist due to belonging to one group or another. The graph in Figure 1 shows the percentage  Detectab 46.94 8.67 57.30 11.68 0.00000 of the value of each variable according to the group. For example, it N can be seen that in the first graph, the D.AT variable (attention deficit) Mann–Whitney U Test (for not normal distribution variables) is positive (value 2) in 30% of the control group and over 70% in the  Omission 46.54 3.60 55.28 13.85 0.00000* group of diagnosed patients. In this and the inattentiveness (INAT) N variable, the differences are marked and reflected in the subsequent  Commiss 48.64 8.76 57.11 12.18 0.00000* Chi-Square test of independence. N Box plots show for continuous variables the differences between quartiles between the different groups, including outliers. It is  Persever 48.63 5.93 58.81 16.25 0.00045* interesting to note the differences in some variables, although their N significance was calculated using an independence test, such as  HRT N 47.33 12.22 53.36 13.24 0.00011* omission (Omission N), perseverance (Persever N) or variability  HRT SD 46.54 7.56 61.65 14.70 0.00000* (Variabil N), are very evident in Figure 2. N Spearman’s rho value is calculated for the continuous variables  Variabil N 45.01 6.48 55.61 11.73 0.00000* to define the graph illustrating the relationship between them. Each  HRT Blo 49.76 8.30 48.03 13.38 0.457403* edge of the graph will correspond to a value greater than 0.5 from CN Table 4. The p-value is added to check the significance of the rho value.  HRT ISI 50.42 9.04 50.63 14.23 0.949235* As a result, three variables are found to be unrelated. In Ch N comparison, the five variables on the left of the graph show a complete Mean values according to the type of patient in the P group (CFS patient) or the Control regular graph as a pentagon, i.e., all five variables are related. The group. The independence test was performed using Student’s t-test if the values correspond to a normal distribution or Mann–Whitney U test otherwise and marked with an asterisk thicker the edge, the higher the correlation, and the thinner and “*.” For values with a p-value of less than 0.05, the hypothesis that group membership redder the values close to 0.5. predetermines the outcome of the variable studied is considered to be accepted. TABLE 3 Chi-square test of independence results. Features importance random forest based Feature Chi square p-Value Cramer’s* The classification algorithm is run in a Python environment value (v 3.7.14), and the library used is sklearn (v 1.1). Random forest Attention deficit 43.14 0.00000 0.44 is a Supervised Machine Learning Algorithm that is used widely Inattentiveness 50.45 0.00000 0.47 Impulsivity 6.27 0.01225 0.17 TABLE 1 Sample analysis. Sustained attention 3.87 0.04920 0.13 C (Control group) P (CFS Diagnosed Vigilance 12.33 0.00045 0.23 group) Gender 6.00 0.01430 0.16 n (225) 67 158 *Attention deficit, inattentiveness, impulsivity, sustained attention, and vigilance was all Age mean (std) 44.82 ± 14.26 51.40 ± 8.11 significant. Gender was also found to be significant. Cramer, greater than 0.30, in Inattentiveness (0.47 Cramer’s *), Attention Deficit (0.43 Cramer’s *), and Vigilance (0.23 Females 53 (79.1%) 145 (91.77%) Cramer’s *) are significant (medium relation if Cramer >0.30 with df = 1). Significant differences indicate that the samples come from different populations. Consequently, it can Males 14 (20.9%) 13 (8.23%) be inferred that the significant variables predict the group the participant belongs to. Frontiers in Psychology 05 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 FIGURE 1 Graph of the contingency tables by groups and normalized, where each bar corresponds to the percentage of each value by groups. Each drawing represents for each dichotomous variable the group differences, with C being the control group and P the diagnosed group. The following acronyms mean the six dichotomous variables: DAT, attention deficit; INAT, inattentiveness; IMPUL, impulsivity; AT.SOST, sustained attention; VIGIL, vigilance; and finally, Gender. features, the most important were HRT N, Detecta N, and A threshold of 60 is identified in the variable Variability N. For the Variability. Both were found to be able to classify ME/CFS values above, it is rare to find patients in the control group, which patients shown in Figure 3. could indicate a risk of cognitive impairment for patients diagnosed Frontiers in Psychology 06 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 FIGURE 2 Box plot numerical features analysis by groups. Box plots for continuous variables by groups. The horizontal axis shows the groups, and the vertical axis shows the variable analyzed. with ME/CFS. The results of the random forests showed that the Figure 4 shows the importance of each variable. HRT N and CPT3™ could discriminate between presentations, mainly for the VARIABILITY N were the critical variables. In contrast, other inattentive presentation. variables related to HRT, such as HRT by Block and HRT ISI Frontiers in Psychology 07 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 TABLE 4 Rho values > abs(0.5) for variables with high correlation in group P (diagnosed ME/CFS). standard” for diagnosing ME/CFS. To the best of our knowledge, this is the first study assessing cognitive dysfunction in ME/CFS Features Rho p-Value with CPT3™. It has been found that CPT3™ could detect Detectab N Commiss N 0.820382 0.00000000 cognitive impairments in all their main attentional measures HRT SD N Variabil N 0.786083 0.00000000 (inattentiveness, impulsivity, sustained attention, and vigilance) in our ME/CFS individuals compared to healthy controls. More Detectab N Omission N 0.776174 0.00000000 specifically, patients had a worse indicator of inattention, Omission N Variabil N 0.766189 0.00000000 sustained attention, vigilance, impulsivity, and slow reaction time, Res.Styl N HRT N 0.722694 0.00000000 showing the likelihood of having an ADHD-like pattern of Omission N HRT SD N 0.691635 0.00000000 functioning. In addition, relevant correlations were obtained between CPT3™ measures in this ME/CFS group. Accordingly, Detectab N Variabil N 0.689658 0.00000000 the adoption of the CPT3™ is supported as a specific test in ME/ Detectab N Persever N 0.673743 0.00000000 CFS aimed at analyzing the cognitive domains (attention and Detectab N HRT SD N 0.614872 0.00000000 response speed) that seem to be key in the cognitive dysfunction Omission N Persever N 0.609529 0.00000000 of ME/CFS. This work is in line with another previous study that assessed Persever N Variabil N 0.590245 0.00000000 different types of attentional impairments in ME/CFS but using Persever N HRT SD N 0.531106 0.00000000 another brief and simple instrument (The Toulouse-Piéron Test) Res.Styl N Omission N 0.521658 0.00000000 that allows the measurement of maintained attention— HRT N HRT SD N 0.501458 0.00000000 concentration, and resistance to monotony, and also, evaluates the Res.Styl N Commiss N −0.563669 0.00000000 multidimensional domain of attention, classified into different types such as arousal attention (alertness/activation), focused attention The undirected graph is constructed with the data from Table 4 using the rho value itself as the weight on each edge, and the result is shown in Figure 2. (detection of a stimulus) and sustained attention (attention to a stimulus or task for a prolonged time). These types of attention problems compromise various neuroanatomic structures, pathways, neurotransmitters, and their receptors. Their results support the reliability of maintained attention as a biomarker of ME/CFS, and attention deficit is a significant disability in patients affected by central fatigue. This neurocognitive dysfunction points to the neural networks involved in attention and focuses the pathological substrate in areas like the anterior cingulate cortex, lateral ventral prefrontal cortex, basal ganglia, or locus coeruleus (Murga et al., 2021). Our results are in agreement with other neurocognitive research in ME/CFS reflecting attentional impairments on cognitive performance in these patients (Murga et al., 2021). Following these current findings, it would be quite interesting to monitor these brain areas with advanced techniques, such as fMRI, PET-scan, etc., to establish the key neurological bases that could be involved in CFS. A neurocognitive profile in ME/CFS has not been described only because of the heterogeneity of the symptoms, but also for FIGURE 3 other shortcomings such as the lack of specific neuropsychological Network relationship between numerical features. The graph represents each continuous variable as a node, indicating the size of tests to evaluate it. CPT3™ represents a reliable alternative for the node as a function of degree (the number of edges). Both the assessing attention disorders in ME/CFS and allows a comprehensive size and color depend on Spearman’s rho value. measurement of multiple cognitive variables quickly. As a computerized test in the form of a computer game, the administration is more practical and stimulating for these patients, Change, had less importance in discriminating ME/ and the elaboration and quantification of scores are accurate. The CFS presentation. Conners Continuous Performance Test (CPT) is widely used in Moreover, the results of the CPT3™ showed that a relevant clinical practice for its usefulness in the study of attention in various number of ME/CFS patients were associated with a likelihood of pathologies such as Attention Deficit Hyperactivity Disorder having a disorder characterized by attention deficits, such as ADHD. (ADHD; Newcorn et al., 2001; Baggio et al., 2020), although a previous study reported that the CPT3™ might be sensitive only to some of the core deficits of ADHD, but not hyperactivity. This Discussion instrument is not considered specific for ADHD, and although CPT may not differentiate between psychiatric and neurological Together with comprehensive diagnostic interviews, disorders that result in executive dysfunctions (Baggio et al., 2020), neuropsychological assessments could give the patient the “gold it can be used in other diseases. Frontiers in Psychology 08 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 FIGURE 4 Scatter plot showing the discriminant threshold. Scatter plot of Variabil N (vertical axe) and HRT N (horizontal axe) where number 1 (red color) corresponds to P group, and number 0 (green color) corresponds to C group. Indicators that measure response speed act as good discriminants CPT3™ makes corrections for age and gender, so published between both groups. Hit Reaction Time (HRT) is the average results have already been corrected for biases identified in several response speed of correct responses for the entire administration, studies (Newcorn et al., 2001; Conners et al., 2003; Seidman et al., measured in milliseconds, for all non-perseverative responses made 2005; Burton et al., 2010; Ramtekkar et al., 2010). T-scores are during the entire administration. An atypically slow HRT may relative to age group and gender, with sex-independent consultation indicate inattention, especially when error rates are high, but may being possible as an option. Although the study design did not take also result from a very conservative response style. Variability, like into account a gender or age balance in the sample, and this should HRT SD, is a measure of response speed consistency; however, be taken into account in future studies, the gender variable was Variability is a “within-respondent” measure (i.e., the amount of surprisingly significant in the Chi-square test. Furthermore, despite variability the respondent exhibited in 18 separate sub-blocks of the the fact that there is also a gender imbalance, especially in the group administration relative to their overall HRT SD score). Although of diagnosed patients, the scores are in line with other studies that Variability is a different measure than HRT SD, the two measures report on the influence of gender on chronic fatigue (Lim typically produce comparable results, and both are related to et al., 2020). inattention. High response speed variability indicates that the CFS/ME patients have more difficulty discriminating between respondent’s attention and processing efficiency varied throughout targets and non-targets, and this poor detectability indicates inattention. the administration. These scores mean that the response speed in the A very unusual number of omission errors may indicate clinical ME/CFS significantly differed from the response speed in the control impairment, fatigue, poor understanding of instructions, or a lack of group. The high variability of the response speed and the slow motivation to respond with full effort. The results of the Mann–Whitney reaction time obtained indicate that ME/CFS suffer dysfunctions in U test show that all variables are significant except HRT Block Change the efficiency of information processing and commit more errors of (HRT Blo C, meaning that the slope of HRT change in the six test omission, commission, and perseverance than healthy controls, blocks) and HRT ISI change [HRT ISI Ch N; indicates that the HRT presenting not only more attention problems, but being slower and change’s slope in the three ISIs (1, 2, and 4 s)]. A T-score of 60–69 in HRT with greater inconsistency in response speed when performing this is classified as slow response, and a T-score of 60–69 in Variability is test. The high variability of response speed indicates that the interpreted as under-average performance. Slower reaction times and attention and processing efficiency of the ME/CFS patients varied high inconsistency of response speed may be associated with the throughout the administration. Figure 5 suggests that this is a inattentive profile. High response speed variability indicates that the discriminant element in differentiating ME/CFS from healthy respondent’s attention and information processing efficiency varied people. The Hit Reaction Time Standard Deviation (HRT SD) throughout the administration (Kao and Thomas, 2010). measures the consistency of the speed of response to target stimuli Frontiers in Psychology 09 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 across the administration. A high HRT SD in CFS/ME patients also be considered discriminant indicators of ME/CFS patients. This tool indicates a greater inconsistency in response speed. It is sometimes makes it possible to assess the severity of cognitive dysfunction in this indicative of inattention, which could suggest that ME/CFS patients population and could improve the diagnosis and/or monitor were less engaged and processed stimuli less efficiently than healthy these patients. subjects during some parts of the CPT3™ administration. Overall, the It should be noted that for a proper interpretation of CPT3™ test ME/CFS group has more-atypical T-scores, which is associated with results in patients diagnosed with CFS, it would be necessary to a high likelihood of having a disorder characterized by attention investigate the clinical history further to assess the individualized deficits, such as ADHD. However, assessors should keep in mind performance of each measured dimension. The impact of medication that other psychological and/or neurological conditions with use on cognitive functioning has not been explored in-depth, and it attention-disrupting symptoms may also generate atypical scores would be interesting to consider it in future studies. Even though the (and thus a high or very high probability estimate). A previous study CPT lacks ecological validity (Baggio et al., 2020) as it does not suggested that ADHD may be common in ME/CFS patients and is adequately simulate the difficulties patients may experience in their associated with a more severe psychopathology clinical profile (Sáez- daily life (i.e., it is free of external distractions that are likely to impair Francàs et al., 2012). the patient’s real-life performance and it is a rather short task to For all the above, CPT3™ represents a reliable alternative to represent overall performance), ME/CFS patients presented altered objectively and comprehensively assess attentional deficits in CPT3™ parameters when compared to healthy people. Most patients CFS. However, it also allows the assessment of other executive were not referred from primary care but belonged to a specialized unit function domains, such as information processing speed and in a tertiary care hospital. They probably have more mental and inhibition, providing the reaction time measures which can non-mental comorbidities, more severe symptoms, and more years of FIGURE 5 Pair plot analysis of the most significant features to discriminate between groups. Histograms as a function of the group and scatter plots of the combinations between the three chosen variables. Frontiers in Psychology 10 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 diagnosis. It should be considered to analyze the relationship between functioning of these patients, and even this computerized test could the results obtained through the CPT3™ with other variables, such as be considered as a possible candidate marker for ME/CFS. As this the severity of symptoms, associated comorbidities, and the time of is the only study reporting CPT3™ scores in adult ME/CFS patients, diagnosis of ME/CFS, evaluating the changes over time in these future studies are needed to compare this test across ME/ participants. Using the test–retest after an estimated suitable time CFS centers. could be convenient for contemplating the evolution of cognitive functioning in these patients. Furthermore, it would be very interesting to analyze the relationship between the degree of fatigue, Data availability statement the psychopathological symptoms, and the cognitive performance of the CPT3™ in these patients. The raw data supporting the conclusions of this article will Finally, it would be highly recommended in future studies to be made available by the authors, without undue reservation. contemplate an analysis of how the presence of comorbid psychopathological symptoms in ME/CFS influences the cognitive performance obtained in the CPT3™, since it is a test that does not Ethics statement discriminate the presence of these factors. Furthermore, it would be interesting to replicate this study by including another group with The studies involving human participants were reviewed and severe psychiatric disorders like major depressive disorder. approved by Ethics Committee of the Vall d’Hebron University Hospital with the registration code PR (AG)257/2021. The patients/ participants provided their written informed consent to participate in Limitations this study. All the patients come from the specialized chronic fatigue unit of the Vall D’Hebrón University Hospital in Barcelona. Patient data from Author contributions other CFS-specialized units with similar protocols could enrich this work. Further research is needed to determine differences in ME/CFS All authors have participated in the conception and design of the patients to validate this work. CPT3™ response values have also not study, and in the collection, analysis and interpretation of the data, as been evaluated based on controlled medication. There is no well as in the writing, revision and approval of the submitted manuscript. information on the influence of COVID-19 infection on the results in either group. Conflict of interest Conclusion JR-Q was on the speakers’ bureau and/or acted as a consultant for Janssen-Cilag, Novartis, Shire, Takeda, Bial, Shionogi, Sincrolab, CPT3™ adequately identifies ME/CFS in this clinical sample of Novartis, BMS, Medicine, and Rubió Uriach, Technofarma, and Raffo adult participants compared to healthy controls. This profile of in the last 3 years. He also received travel awards (air tickets + hotel) cognitive dysfunction could be related to other pathophysiological for taking part in psychiatric meetings from Janssen-Cilag, Rubió phenomena of CFS, and its determination could be key to elucidating Shire, Takeda, Shionogi, Bial, and Medice. The Department of the underlying basis and providing empirical evidence for the Psychiatry chaired by him received unrestricted educational and usefulness of this computerized neuropsychological test. Additionally, research support from the following companies in the last 3 years: fatigue is the most common symptom associated with NCDs. There is Janssen-Cilag, Shire, Oryzon, Roche, Psious, and Rubió AR-U has insufficient evidence examining the relationship between fatigue and received travel awards (air tickets + hotel) for taking part in annual cognitive impairments in patients with other chronic diseases in psychiatric meetings from Lundbeck and has acted as a speaker at which the symptom fatigue appears (e.g., major depressive disorder, various training courses financed by Organon, Janssen-Cilag, fibromyalgia, postcovida, etc.). Further studies using CPT3™ could and Lundbeck. be useful to examine possible relationships between these The remaining authors declare that the research was conducted in two symptoms. the absence of any commercial or financial relationships that could This study shows that a relevant number of ME/CFS patients be construed as a potential conflict of interest. had CPT3™ values compatible with a likelihood of having a disorder characterized by attention deficits, suggesting the possible existence of an underlying neurological basis for attentional dysfunction Publisher’s note among the study patients. Furthermore, it was shown that two measures of response speed (hit reaction time and variability) act All claims expressed in this article are solely those of the authors as good discriminants between both groups. Taken together, CPT3™ and do not necessarily represent those of their affiliated is a helpful neuropsychological instrument for discriminating organizations, or those of the publisher, the editors and the cognitive impairments in attention and response speed in reviewers. Any product that may be evaluated in this article, or CFS. Therefore, the results obtained here will allow us to justify the claim that may be made by its manufacturer, is not guaranteed or use of the CPT3™ in other investigations that explore the cognitive endorsed by the publisher. Frontiers in Psychology 11 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 References Afari, N., and Buchwald, D. (2003). Chronic fatigue syndrome: a review. Am. J. Hagberg, A., Swart, P. S., and Chult, D. (2008) Exploring network structure, dynamics, Psychiatry 160, 221–236. doi: 10.1176/appi.ajp.160.2.221 and function using NetworkX. Los Alamos, NM (United States): Los Alamos National Lab (LANL). Alonso, J., Prieto, L., and Antó, J. M. (1995). The Spanish version of the SF-36 health survey (the SF-36 health questionnaire): an instrument for measuring clinical results. Kao, G. S., and Thomas, H. M. (2010). Test review: C. Keith Conners Conners 3rd Med. Clin. 104, 771–776. edition Toronto, Ontario, Canada: multi-health systems, 2008. J. Psychoeduc. Assess. 28, 598–602. doi: 10.1177/0734282909360011 Badin, O. O., Smith, M. R., Conte, D., and Coutts, A. J. (2016). Mental fatigue: impairment of technical performance in small-sided soccer games. Int. J. Sports Physiol. Kim, H.-Y. (2017). Statistical notes for clinical researchers: chi-squared test and Perform. 11, 1100–1105. doi: 10.1123/ijspp.2015-0710 Fisher’s exact test. Restor. Dent. Endod. 42, 152–155. doi: 10.5395/rde.2017.42.2.152 Baggio, S., Hasler, R., Giacomini, V., El-Masri, H., Weibel, S., Perroud, N., et al. (2020). Liaw, A., and Wiener, M. (n.d.). Classification and regression by randomForest. R news Does the continuous performance test predict ADHD symptoms severity and ADHD [Preprint]. Available at: https://www.researchgate.net/file.PostFileLoader.html?id=57d98 presentation in adults? J. Atten. Disord., 840–848. doi: 10.1177/1087054718822060 4d196b7e421f8398993&assetKey=AS%3A406267404275720%401473873105695 (Accessed February 13, 2023). Bassi, N., Amital, D., Amital, H., Doria, A., and Shoenfeld, Y. (2008). Chronic fatigue syndrome: characteristics and possible causes for its pathogenesis. Israel Med. Assoc. J. Lim, E.-J., Ahn, Y. C., Jang, E. S., Lee, S. W., Lee, S. H., and Son, C. G. (2020). 10, 79–82. Systematic review and meta-analysis of the prevalence of chronic fatigue syndrome/ myalgic encephalomyelitis (CFS/ME). J. Transl. Med. 18:100. doi: 10.1186/ Beaumont, A., Burton, A. R., Lemon, J., Bennett, B. K., Lloyd, A., and s12967-020-02269-0 Vollmer-Conna, U. (2012). Reduced cardiac vagal modulation impacts on cognitive performance in chronic fatigue syndrome. PLoS One 7:e49518. doi: 10.1371/journal. McGregor, N. R., Dunstan, R. H., Butt, H. L., Roberts, T. K., Klineberg, I. J., and pone.0049518 Zerbes, M. (1997). A preliminary assessment of the association of SCL-90-R psychological inventory responses with changes in urinary metabolites in patients with Berger, I., Slobodin, O., and Cassuto, H. (2017). Usefulness and validity of continuous chronic fatigue syndrome. J. Chronic Fatigue Syndr. 3, 17–37. doi: 10.1300/ performance tests in the diagnosis of attention-deficit hyperactivity disorder children. J092v03n01_03 Arch. Clin. Neuropsychol. 32, 81–93. doi: 10.1093/arclin/acw101 Menzies, V., Kelly, D. L., Yang, G. S., Starkweather, A., and Lyon, D. E. (2021). A Burton, L., et al. (2010). Effects of gender and personality on the Conners continuous systematic review of the association between fatigue and cognition in chronic performance test. J. Clin. Exp. Neuropsychol. 32, 66–70. doi: 10.1080/13803390902806568 noncommunicable diseases. Chronic Illn. 17, 129–150. doi: 10.1177/1742395319836472 Carruthers, B. M., Jain, A. K., de Meirleir, K. L., Peterson, D. L., Klimas, N. G., Morrow, S. A., Rosehart, H., and Johnson, A. M. (2015). ‘Diagnosis and quantification Lerner, A. M., et al. (2003). Myalgic encephalomyelitis/chronic fatigue syndrome. J. of cognitive fatigue in multiple sclerosis’, cognitive and behavioral neurology: official Chron. Fatig. Syndr. 11, 7–115. doi: 10.1300/j092v11n01_02 journal of the Society for Behavioral and Cognitive. Neurology 28, 27–32. doi: 10.1097/ Carruthers, B. M., van de Sande, M. I., de Meirleir, K. L., Klimas, N. G., Broderick, G., WNN.0000000000000050 Mitchell, T., et al. (2011). Myalgic encephalomyelitis: international consensus criteria. J. Murga, I., Aranburu, L., Gargiulo, P. A., Gómez-Esteban, J. C., and Lafuente, J. V. Intern. Med. 270, 327–338. doi: 10.1111/j.1365-2796.2011.02428.x (2021). The maintained attention assessment in patients affected by Myalgic Castresana, C., Pérez, A. G.-E., and de Rivera, J. L. G. (1995). Hospital anxiety and encephalomyelitis/chronic fatigue syndrome: a reliable biomarker? J. Transl. Med. depression scale y psicopatología afectiva. Anal. Psiquiatr. 2, 126–130. 19:494. doi: 10.1186/s12967-021-03153-1 Castro-Marrero, J., Sáez-Francàs, N., Segundo, M. J., Calvo, N., Faro, M., Aliste, L., Nacul, L. C., Lacerda, E. M., Pheby, D., Campion, P., Molokhia, M., Fayyaz, S., et al. et al. (2016). Effect of coenzyme Q10 plus nicotinamide adenine dinucleotide (2011). Prevalence of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) supplementation on maximum heart rate after exercise testing in chronic fatigue in three regions of England: a repeated cross-sectional study in primary care. BMC Med. syndrome – a randomized, controlled, double-blind trial. Clin. Nutr. 35, 826–834. doi: 9:91. doi: 10.1186/1741-7015-9-91 10.1016/j.clnu.2015.07.010 Newcorn, J. H., Halperin, J. M., Jensen, P. S., Abikoff, H. B., Arnold, L. E., Castro-Marrero, J., Sáez-Francàs, N., Santillo, D., and Alegre, J. (2017). Treatment and Cantwell, D. P., et al. (2001). Symptom profiles in children with ADHD: effects of management of chronic fatigue syndrome/myalgic encephalomyelitis: all roads lead to comorbidity and gender. J. Am. Acad. Child Adolesc. Psychiatry 40, 137–146. doi: Rome. Br. J. Pharmacol. 174, 345–369. doi: 10.1111/bph.13702 10.1097/00004583-200102000-00008 Cockshell, S. J., and Mathias, J. L. (2010). Cognitive functioning in chronic fatigue Newman, M. (2018). Networks. 2nd Edn. UK: Oxford University Press. syndrome: a meta-analysis. Psychol. Med. 40, 1253–1267. doi: 10.1017/ Prinsen, H., van Dijk, J. P., Zwarts, M. J., Leer, J. W. H., Bleijenberg, G., and van S0033291709992054 Laarhoven, H. W. M. (2015). The role of central and peripheral muscle fatigue in Conners, C. K., Epstein, J. N., Angold, A., and Klaric, J. (2003). Continuous postcancer fatigue: a randomized controlled trial. J. Pain Symptom Manag. 49, 173–182. performance test performance in a normative epidemiological sample. J. Abnorm. Child doi: 10.1016/j.jpainsymman.2014.06.020 Psychol. 31, 555–562. doi: 10.1023/a:1025457300409 Prins, J. B., van der Meer, J. W. M., and Bleijenberg, G. (2006). Chronic fatigue Cvejic, E., Birch, R. C., and Vollmer-Conna, U. (2016). Cognitive dysfunction in syndrome. Lancet 367, 346–355. doi: 10.1016/s0140-6736(06)68073-2 chronic fatigue syndrome: a review of recent evidence. Curr. Rheumatol. Rep. 18:24. doi: Ramtekkar, U. P., Reiersen, A. M., Todorov, A. A., and Todd, R. D. (2010). Sex and age 10.1007/s11926-016-0577-9 differences in attention-deficit/hyperactivity disorder symptoms and diagnoses: Deluca, J., Christodoulou, C., Diamond, B. J., Rosenstein, E. D., Kramer, N., and implications for DSM-V and ICD-11. J. Am. Acad. Child Adolesc. Psychiatry 49, Natelson, B. H. (2004). Working memory deficits in chronic fatigue syndrome: 217–28.e1–3. doi: 10.1016/j.jaac.2009.11.011 differentiating between speed and accuracy of information processing. J. Int. Rasouli, O., Gotaas, M. E., Stensdotter, A. K., Skovlund, E., Landrø, N. I., Dåstøl, P., Neuropsychol. Soc. 10, 101–109. doi: 10.1017/S1355617704101124 et al. (2019). Neuropsychological dysfunction in chronic fatigue syndrome and the Dickson, A., Toft, A., and O’Carroll, R. E. (2009). Neuropsychological functioning, relation between objective and subjective findings. Neuropsychology 33, 658–669. doi: illness perception, mood and quality of life in chronic fatigue syndrome, autoimmune 10.1037/neu0000550 thyroid disease and healthy participants. Psychol. Med. 39, 1567–1576. doi: 10.1017/ Reyes, M., Nisenbaum, R., Hoaglin, D. C., Unger, E. R., Emmons, C., Randall, B., et al. S0033291708004960 (2003). Prevalence and incidence of chronic fatigue syndrome in Wichita, Kansas. Arch. Doob, J. L. (1946). ‘Mathematical Methods of Statistics. Harald Cramèr. Princeton, NJ: Intern. Med. 163, 1530–1536. doi: 10.1001/archinte.163.13.1530 Princeton Univ. Press, 1946. Pp. xvi 575. (Illustrated.) $6.00’, Science, 450. Rivera, J., Alegre, C., Ballina, F. J., Carbonell, J., Carmona, L., Castel, B., et al. (2006). Fasmer, O. B., Mjeldheim, K., Førland, W., Hansen, A. L., Syrstad, V. E. G., Documento de consenso de la Sociedad Española de Reumatología sobre la fibromialgia. Oedegaard, K. J., et al. (2016). Linear and non-linear analyses of Conner’s continuous Reumatol. Clín. 2, S55–S66. doi: 10.1016/s1699-258x(06)73084-4 performance test-II discriminate adult patients with attention deficit hyperactivity Ruiz, E., Alegre, J., García Quintana, A. M., Aliste, L., Blázquez, A., and Fernández de disorder from patients with mood and anxiety disorders. BMC Psychiatry 16:284. doi: Sevilla, T. (2011). Síndrome de fatiga crónica: estudio de una serie consecutiva de 824 10.1186/s12888-016-0993-4 casos evaluados en dos unidades especializadas. Nuevos Arch. Facult. Med. 211, 385–390. Fisk, J. D., Ritvo, P. G., Ross, L., Haase, D. A., Marrie, T. J., and Schlech, W. F. (1994). doi: 10.1016/j.rce.2011.02.013 Measuring the functional impact of fatigue: initial validation of the fatigue impact scale. Sáez-Francàs, N., Alegre, J., Calvo, N., Antonio Ramos-Quiroga, J., Ruiz, E., Clin. Infect. Dis. 18, S79–S83. doi: 10.1093/clinids/18.supplement_1.s79 Hernández-Vara, J., et al. (2012). Attention-deficit hyperactivity disorder in chronic Fisk, J. D., and Doble, S. E. (2002). Construction and validation of a fatigue impact fatigue syndrome patients. Psychiatry Res. 200, 748–753. doi: 10.1016/j. scale for daily administration (D-FIS). Qual. Life Res. 11, 263–272. doi: psychres.2012.04.041 10.1023/a:1015295106602 Santamarina-Perez, P., Eiroa-Orosa, F. J., Rodriguez-Urrutia, A., Qureshi, A., and Fukuda, K., Straus, S. E., Hickie, I., Sharpe, M. C., Dobbins, J. G., and Komaroff, A. Alegre, J. (2014). Neuropsychological impairment in female patients with chronic fatigue (1994). The chronic fatigue syndrome: a comprehensive approach to its definition and syndrome: A preliminary study. Appl. Neuropsychol. 21, 120–127. doi: study. Ann. Intern. Med. 121, 953–959. doi: 10.7326/0003-4819-121-12-199412150-00009 10.1080/09084282.2013.771264 Frontiers in Psychology 12 frontiersin.org Fernández-Quirós et al. 10.3389/fpsyg.2023.1127193 Seidman, L. J., Biederman, J., Monuteaux, M. C., Valera, E., Doyle, A. E., and UniversityProQuest Dissertations Publishing, Degree Year, p. 3007026. Available at: Faraone, S. V. (2005). Impact of gender and age on executive functioning: do girls and https://search.proquest.com/openview/5d1cbf13c930b7358050381ebab41a85/1?pq-ori boys with and without attention deficit hyperactivity disorder differ neuropsychologically gsite=gscholar&cbl=18750&diss=y (Accessed February 13, 2023). in preteen and teenage years? Dev. Neuropsychol. 27, 79–105. doi: 10.1207/ Swain, M. G. (2000). Fatigue in chronic disease. Clin. Sci. 99, 1–8. doi: 10.1042/ s15326942dn2701_4 CS19990372 Suchowski, M. A. (n.d.). An analysis of the impact of an outlier on correlation Verster, J. C., Pandi-Perumal, S. R., and Streiner, D. L. (2008). Sleep and Quality of Life coefficients across small sample data where rho is non –zero. Western Michigan in Clinical Medicine Springer Science & Business Media. Frontiers in Psychology 13 frontiersin.org