Psychological Medicine Polygenic association between attention- cambridge.org/psm deficit/hyperactivity disorder liability and cognitive impairments Isabella Vainieri1 , Joanna Martin2, Anna-Sophie Rommel3, Philip Asherson1, Original Article Tobias Banaschewski4, Jan Buitelaar5,6, Bru Cormand7,8,9,10, Jennifer Crosbie11, Cite this article: Vainieri I et al (2022). Polygenic association between attention- Stephen V. Faraone12, Barbara Franke13, Sandra K. Loo14, Ana Miranda15, deficit/hyperactivity disorder liability and cognitive impairments. Psychological Medicine Iris Manor16, Robert D. Oades17, Kirstin L. Purves1, 52, 3150–3158. https://doi.org/10.1017/ S0033291720005218 J. Antoni Ramos-Quiroga18,19,20,21, Marta Ribasés18,19,20,22, Herbert Roeyers23, Aribert Rothenberger24, Russell Schachar11, Joseph Sergeant25, Received: 16 April 2020 Revised: 15 July 2020 Hans-Christoph Steinhausen26,27,28,29, Pieter J. Vuijk30, Alysa E. Doyle31 Accepted: 8 December 2020 First published online: 3 February 2021 and Jonna Kuntsi1 Key words: ADHD; attention; cognition; inhibition; polygenic risk scores; reaction time variability Abstract Background. A recent genome-wide association study (GWAS) identified 12 independent loci Author for correspondence: Jonna Kuntsi, E-mail: jonna.kuntsi@kcl.ac.uk significantly associated with attention-deficit/hyperactivity disorder (ADHD). Polygenic risk scores (PRS), derived from the GWAS, can be used to assess genetic overlap between ADHD and other traits. Using ADHD samples from several international sites, we derived PRS for ADHD from the recent GWAS to test whether genetic variants that contribute to ADHD also influence two cognitive functions that show strong association with ADHD: attention regulation and response inhibition, captured by reaction time variability (RTV) and commission errors (CE). Methods. The discovery GWAS included 19 099 ADHD cases and 34 194 control participants. The combined target sample included 845 people with ADHD (age: 8–40 years). RTV and CE were available from reaction time and response inhibition tasks. ADHD PRS were calculated from the GWAS using a leave-one-study-out approach. Regression analyses were run to investigate whether ADHD PRS were associated with CE and RTV. Results across sites were combined via random effect meta-analyses. Results. When combining the studies in meta-analyses, results were significant for RTV (R2 = 0.011, β = 0.088, p = 0.02) but not for CE (R2 = 0.011, β = 0.013, p = 0.732). No significant association was found between ADHD PRS and RTV or CE in any sample individually ( p > 0.10). Conclusions. We detected a significant association between PRS for ADHD and RTV (but not CE) in individuals with ADHD, suggesting that common genetic risk variants for ADHD influence attention regulation. Introduction A recent case-control genome-wide association study (GWAS) identified, for the first time, 12 independent loci significantly associated with attention-deficit/hyperactivity disorder (ADHD) (Demontis et al., 2019). This GWAS enables further genetic investigations using polygenic risk scores (PRS), which are calculated for each individual by computing the sum of their risk alleles across the genome, weighted by effect sizes (Choi, Mak, & O’Reilly, 2018). PRS provide an estimate of the genetic propensity to ADHD at the individual level that can be used to investigate shared genetic etiology between ADHD and other phenotypes. Previous studies on general population samples show that ADHD PRS are associated with a wide range of psychiatric and somatic disorders and traits, such as depression, anxiety, neur- © The Author(s), 2021. Published by oticism, irritability, childhood internalizing and externalizing symptoms, obesity-related phe- Cambridge University Press. This is an Open notypes and smoking (Brikell et al., 2018; Du Rietz et al., 2018; Riglin et al., 2017). Only a few Access article, distributed under the terms of of these population-based studies explored the cognitive phenotypes associated with ADHD the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), using polygenic approaches, but have provided initial evidence for an association between which permits unrestricted re-use, PRS for ADHD and lower general cognitive ability (Du Rietz et al., 2018; Martin, distribution, and reproduction in any medium, Hamshere, Stergiakouli, O’Donovan, & Thapar, 2015a), educational attainment (Stergiakouli provided the original work is properly cited. et al., 2017) and working memory, but not inhibition impairments (measured with the Opposite Words Task; Martin et al., 2015a). Evidence from clinically diagnosed samples with ADHD remains even more limited. The findings reported to date indicate an association of ADHD PRS with low academic achievement (Vuijk et al., 2019) and poor working memory and arousal-alertness, measured with latent variables (Nigg et al., 2018). In contrast, no Psychological Medicine 3151 significant associations emerged between PRS for ADHD and approach: International Multisite ADHD Genetics Project latent variables capturing inhibition or speed of responses (Nigg (IMAGE-I, subdivided here to IMAGE-8 and IMAGE-Dutch et al., 2018). A recent study found that PRS for ADHD were asso- that had different cognitive test batteries), University of ciated with a measure of interference, the ‘variance of word inter- California Los Angeles (UCLA), Toronto and Barcelona. All par- ference time’ in the Stroop test (Chang, Yang, Wang, & Faraone, ticipants for each site completed a comprehensive protocol of cog- 2020). nitive tasks, which differed for each site. Participants from We now extend, in a sample of 845 people with ADHD, the IMAGE-8 performed a four-choice reaction time task (Fast previous PRS investigations of ADHD-related cognitive pheno- task) and a version of the Go/No-Go task with fast and slow con- types to two cognitive measures that have extensive evidence ditions, whereas IMAGE-Dutch participants performed the from phenotypic studies of a strong association with ADHD, Stop-Signal Task (SST). At UCLA and Barcelona, participants but have not yet been investigated using PRS: increased reaction performed the Continuous Performance Test II (CPT-II), whereas time variability (RTV) and commission errors (CE) (Kuntsi the Go/No-Go task was administered in Toronto. Descriptive sta- et al., 2010; Loo et al., 2009; Schachar et al., 2007; van Rooij tistics for each sample are shown in Table 1. Based on previous et al., 2015). RTV captures the highly variable speed of respond- publications, cognitive variables were selected from the tasks ing that is strongly characteristic of people with ADHD across a that showed a significant ADHD case-control difference (effect variety of cognitive tasks requiring a fast response (Kofler et al., sizes ranging from 0.32 to 0.95 for RTV, and from 0.38 to 0.42 2013; Kuntsi et al., 2013), and has been linked in EEG and skin for CE; Alemany et al., 2015; Hale et al., 2014; Kuntsi et al., conductance studies to attention allocation and peripheral 2010; Schachar et al., 2007; van Rooij et al., 2015). RTV [standard hypo-arousal (Cheung et al., 2017; James, Cheung, Rijsdijk, deviation (S.D.) of reaction times] was obtained from each of the Asherson, & Kuntsi, 2016). CE, which represent the responses tasks. Evidence for comparability between tasks was previously to non-target stimuli on inhibitory tasks such as the Go/No-Go obtained from model fitting analyses on the fast task and Go/ task, capture failures to withhold responding. No-Go task, which indicated a high familial correlation (rf = Family and twin studies suggest a significant degree of familial/ 0.75) between RTVs obtained from each task, suggesting they genetic sharing between ADHD and both RTV and CE (Kuntsi are measuring largely the same liability (Kuntsi et al., 2010). CE et al., 2010, 2014). For example, in a large study of 1265 children was obtained from the CPT-II and Go/No-Go tasks only. The and adolescents, including 464 participants with ADHD, we high rates of Go-stimuli in the CPT-II task make this task com- observed a familial correlation of 0.74 between ADHD and parable to a Go/No-Go task. RTV, and 0.45 between ADHD and CE (Kuntsi et al., 2010). The analyses further indicated a significant degree of etiological separ- IMAGE-I ation in the association of ADHD with RTV and CE (Kuntsi Sample: IMAGE-I is a European project on ADHD familiality et al., 2010), with a similar conclusion emerging also from model using a common protocol of centralized training and data man- fitting analyses in a population twin sample of 1312 children agement. IMAGE-I includes data from different European sites (Kuntsi et al., 2014). Family model fitting analyses also showed a and Israel, recruited from specialist clinics in Tel-Aviv, Essen, high familial correlation between RTV obtained from two different Gottingen, Brussels, Dublin, Valencia, Zurich, London, tasks (a four-choice reaction time task, the Fast task, and a Go/ Nijmegen and Amsterdam (Kuntsi, Neale, Chen, Faraone, & No-Go task; rf = 0.75) (Kuntsi et al., 2010), suggesting RTV can Asherson, 2006a; Müller et al., 2011a; Müller et al., 2011b). The be combined across such tasks for further genetic investigations. full IMAGE-I sample consisted of 782 individuals with DSM-IV Using a polygenic approach, we can move beyond the inferred ADHD combined type (680 ADHD combined type probands etiological sharing between ADHD and RTV or CE that rely on including 102 of their siblings who also met criteria for ADHD) comparisons of related individuals (in twin and family designs), to and 808 additional unaffected siblings aged 6–19 years (Kuntsi test the associations using molecular genetic data in unrelated indi- et al., 2006a). All participants were recruited from specialist viduals. Specifically, in this collaborative study using ADHD samples clinics. In IMAGE-I, parents of children were interviewed by from several international sites, we derive PRS for ADHD from the recent GWAS (Demontis et al., 2019) to test whether genetic var- iants that contribute to ADHD also influence the cognitive impair- Table 1. Descriptive statistics for all samples ments captured by RTV and CE in people with ADHD. IQ Age Sex Sample N mean (S.D.) mean (S.D.) M:F Methods IMAGE-I Discovery sample IMAGE-8 143 103.78 11.30 231:26 (15.24) (2.67) As the discovery dataset, we used the Psychiatric Genomics Consortium (PGC) and iPSYCH Danish data analyzed in the IMAGE-Dutch 226 98.96 11.50 119:24 recently published GWAS of ADHD (Demontis et al., 2019). (11.52) (2.47) This GWAS consists of 11 studies, with a total of 19 099 UCLA 55 113.33 11.43 30:25 ADHD cases and 34 194 control subjects of European ancestry (15.03) (2.98) (full sample sizes are given in online Supplementary Table S1). Toronto 54 101.24 9.38 42:12 (11.40) (2.12) Barcelona 367 NA 33.24 249:118 Target samples and cognitive assessments (10.54) From the above discovery sample, four sub-samples from different IMAGE, International Multisite ADHD Genetics Project; UCLA, University of California Los sites were used as target samples applying a leave-one-study-out Angeles. 3152 Isabella Vainieri et al. trained researchers with the Parental Account of Childhood were adjusted by 50 ms after every failed or successful response, Symptom (PACS), a semi-structured, standardized, investigator- leading to an approximate 50% success rate on the Stop-trials based interview developed as an instrument to provide an object- for all participants. The task consisted of two practice blocks ive measure of child behavior. Both parents and teachers com- and four test blocks, each consisting of 60 trials. pleted the respective versions of the Conners’ ADHD rating scales and the Strengths and Difficulties Questionnaire (SDQ). UCLA Exclusion criteria were autism, epilepsy, IQ < 70, brain disorders Sample: At UCLA, 156 participants with ADHD were recruited as and any genetic or medical disorder associated with externalizing part of the PUWMa collaboration [Pfizer-funded study from the behaviors that might mimic ADHD. Wherever possible, families University of California, Los Angeles (UCLA), Washington withdrew stimulant medication for 1 week prior to research University, and Massachusetts General Hospital (MGH)], which assessments to allow for more accurate ascertainment of the included 540 children and adolescents aged 5–18 years, and 519 current level of ADHD symptoms and behaviors. Alternatively, of their parents, ascertained from 370 families with clinical interviews were based on medication-free periods. A min- ADHD-affected sibling pairs. Children and adolescents were imum of a 48-h medication-free period was required for cognitive assessed according to DSM-IV-TR criteria. Families were testing. All data were collected with informed consent of recruited through clinical referrals, schools and responses to the parents and with the approval of the site’s Institutional advertisements (e.g. newsletters, community newspapers or flyers Review Board (IRB) or Ethical Committee. distributed at parent meetings in the greater Los Angeles area). Due to differences in the protocol of the cognitive tasks, Respondents without a previous diagnosis of ADHD were IMAGE-I can be subdivided into two subsamples: IMAGE-8 screened with the parent and teacher version of the Swanson, (including participants from Tel-Aviv, Essen, Gottingen, Nolan and Pelham Rating Scale, SNAP-IV (Swanson et al., Brussels, Dublin, Valencia, Zurich and London) and 2012). After initial screening, children and adolescents were IMAGE-Dutch (including participants from Nijmegen and assessed by master’s level clinical psychologists or highly trained Amsterdam). In the current study, we included only participants interviewers using the Schedule for Affective Disorders and with an ADHD diagnosis who had both cognitive and genetic Schizophrenia for School-Age Children-Present and Lifetime data available. The final sample consisted of 143 ADHD partici- version (K-SADS-PL), as well as a parent-completed Child pants from the IMAGE-8 study and 226 ADHD participants Behaviour Checklist (CBCL) and Teacher Report Form. from the IMAGE-Dutch study. Participants were excluded if they were positive for any of the Tasks: Fast-Task, Go/No-Go and SST: The Fast task is a com- following: neurological disorder, head injury resulting in concus- puterized four-choice reaction time (RT) task which measures sion, lifetime diagnoses of schizophrenia or autism or estimated performance under a baseline (slow-unrewarded) and a IQ < 70. Participants on stimulant medication were asked to dis- fast-incentive condition (Andreou et al., 2007; Kuntsi et al., continue use for 24 h prior to their visit. The final sample with 2006b). In the current study, only data from the baseline condi- both cognitive and genetic data available consists of 55 ADHD tion was included as this condition is more sensitive to ADHD cases. (Kuntsi et al., 2013). The baseline condition consisted of 72 trials. Task: CPT-II: The CPT-II (Conners, 2000) is a 14-min compu- Four empty circles (warning signals, arranged horizontally) first terized task that consisted of six blocks and three sub-blocks. appeared for 8 s, after which one of them (the target) was colored Participants were required to press the space button on the key- in. Participants were asked to press the response key that corre- board whenever any letter except the letter ‘X’ appeared on the sponded to the position of the target. Following a response, the computer screen. The task consisted of 360 trials, including 36 stimuli disappeared from the screen and a fixed inter-trial interval presentations of the inhibition target (X). Targets (including of 2.5 s followed. Speed and accuracy were emphasized equally. ‘go’ targets: A, B, C, D, F, I, L, O, T) were presented in rando- The Go/No-Go task is a computerized test used to assess mized order for 250-ms with variable inter-trial interval of 750, inhibitory control (Börger & van der Meere, 2000; Kuntsi, 1750 and 3750 ms. The presentation order of the different Andreou, Ma, Börger, & van der Meere, 2005; van der Meere, inter-trial intervals varied between blocks. The Go:No/Go ratio Stemerdink, & Gunning, 1995). On each trial of the Go/No-Go was 9:1. task, one of two possible stimuli appeared for 300 ms in the mid- dle of the computer screen. The child was instructed to respond only to the Go stimuli and to withhold their response to Barcelona No-Go stimuli. Participants were asked to react as quickly as pos- Sample: The Spanish sample included 607 adults with ADHD sible while maintaining a high level of accuracy. The proportion of (age range 18–40 years), recruited and evaluated at the Hospital Go stimuli to No-Go stimuli was 4:1. This version of the Go/ Universitari Vall d’Hebron in Barcelona. The diagnosis of No-Go task consisted of three conditions (slow, fast and incen- ADHD was evaluated by clinicians with the Structured Clinical tive). Here, we use data only from the slow condition, which Interview for DSM-IV Axis I and II Disorders (SCID-I and show a strong association with ADHD (Andreou et al., 2007; SCID-II) and the Conner’s Adult ADHD Diagnostic Interview Kuntsi, Wood, Van Der Meere, & Asherson, 2009; Uebel et al., for DSM-IV (CAADID Parts I and II). Exclusion criteria were 2010). The slow condition consisted of 72 trials and were pre- IQ < 70, schizophrenia or other psychotic disorders, ADHD sented with a fixed inter-stimulus interval of 8 s. symptoms due to mood, anxiety, dissociative or personality disor- The SST is a response inhibition task, where participants had ders, adoption, sexual or physical abuse, birth weight <1.5 kg and to respond as quickly as possible to a Go stimulus by left or right other neurological or systemic disorders that might explain button press, unless shortly after presentation it was followed by a ADHD symptoms. Cognitive and genetic data were available Stop signal, in which case they were to withhold their response from 367 ADHD participants. More information about this sam- (25% of trials) (Logan, Cowan, & Davis, 1984). The task difficulty ple can be found elsewhere (Sánchez-Mora et al., 2015). was adaptive, meaning delays between the Go and Stop stimulus Task: CPT-II: See UCLA for task description. Psychological Medicine 3153 Toronto to keep a set of independent SNPs. Linear regression models Sample: The initial Canadian ADHD sample included 248 chil- were used to estimate associations between PRS and phenotypes dren aged 6–16 years referred for ADHD, learning and/or behav- in the target samples. PRS were calculated at a number of ioral problems to the Hospital for Sick Children, Toronto (Lionel p value thresholds for SNP inclusion to provide the most pre- et al., 2011). ADHD diagnostic information was obtained based dictive PRS. The p value thresholds used were 0.001, 0.05, 0.1, on DSM-IV criteria from parents and teachers in semi-structured 0.2, 0.3, 0.4, 0.5 and 1. We included age, sex and the first five clinical interviews including the Parent Interview for Child principal components (PCs) as covariates in all analyses, to con- Symptoms (PICS) and the Teacher Telephone Interview (TTI). trol for population stratification. The number of PCs was chosen The assessments were conducted by a social worker, a clinical based on the cohort’s sample size (all <1000) in order to avoid nurse specialist or a clinical psychologist and supervised by a clin- overfitting and to reflect the differential power to capture true ical psychologist or child psychiatrist. Exclusion criteria were an population structure by principal component analysis, as IQ < 80, pervasive developmental disorder, autism or comorbid reported in Demontis et al. (2019). The estimated amount of psychiatric disorder that could better account for the disorder variance explained by PRS (i.e. R 2 values) that we report for Participants who were treated with stimulant medication had to each study are adjusted from a baseline model including the be unmedicated for a minimum of 24 h before assessment and covariates; the reported regression coefficients and standard testing. Cognitive and genetic data for this study were available errors (S.E.) were standardized to have mean = 0 and S.D. 1 from 54 children with ADHD. using the PRSice command (--score std). We performed strin- Task: Go/No-Go task: This version of the Go/No-Go task gent permutation testing within PRSice-2 using 10 000 permu- involved 128 trials of which 32 were No-Go trials and 96 were tations to control for type 1 error and to prevent data overfitting Go trials. During the Go task, one of two possible letters was pre- across the range of p value thresholds considered (0.001, 0.05, sented (an X or an O) on each trial. Participants were required to 0.1, 0.2, 0.3, 0.4, 0.5 and 1). The p values are reported before make a response to the Go task stimuli as quickly and as accur- correction (indicated with ‘p’), and after correction (indicated ately as possible by pressing one key of a handheld response with ‘empirical-p’). Online Supplementary Figs S1–S9 provides box for an X and the other for an O (Go stimuli). The No-Go plots for the PRS prediction models for RTV and CE across all task involved an auditory tone which was presented, at the sites. same time as the stimulus (letters), at random, on 25% of trials. Participants were instructed to withhold their response when Meta-analyses they heard the tone. The Go task stimulus was presented for For the meta-analyses, we used a random effects model using 1000 ms immediately following a fixation point of 500 ms. The the rma.uni function of the metafor package in R, with the task included four blocks, each with 24 Go trials and eight method set to ‘REML’. Meta-analyses for both RTV and CE No-Go Trials. The Go:No-Go ratio was 3:1. were performed across all samples at all thresholds to check the consistency of the associations between PRS and these mea- Data analyses sures (online Supplementary Tables S2 and S3). Combining all samples, the sample size for the meta-analysis consisted of n = Quality control of genetic and cognitive data 743 ADHD participants for RTV and n = 679 ADHD partici- Quality control of genetic data was previously performed and was pants for CE. available for analyses (for more information see Demontis et al., 2019). To account for positive skewness of the cognitive data, we Results applied appropriate transformations to all cognitive measures PRS in individual datasets for each variable prior to analyses. Square root transformations were used in all samples for CE. For RTV, we used a logarithm PRS for ADHD were not significantly associated with RTV in any transformation for IMAGE-8 team, Dutch-IMAGE and UCLA, of the individual datasets (R 2 = 0.004, p = 0.771, empirical-p = 0.993, and square root transformation for Barcelona and Toronto. β = 0.024 for IMAGE-8; R 2 = 0.016, p = 0.124, empirical-p = 0.317, There were no extreme outliers for RTV or CE (>3.5 S.D.). β = 0.135 for IMAGE-Dutch; R 2 = 0.008, p = 0.466, empirical-p = 0.823, β = 0.032 for UCLA; R 2 = 0.031, p = 0.362, empirical-p PRS analyses = 0.459, β = 0.112 for Toronto; R 2 = 0.012, p = 0.029, The GWAS summary statistics used as the discovery sample in- empirical-p = 0.079, β = 0.122 for Barcelona). All associations cluded the four target sub-groups (IMAGE-I, Toronto, Barcelona showed a positive direction. PRS for ADHD were not significantly and UCLA). For this reason, PRS were calculated from the associated and showed inconsistent direction of association with main GWAS each time excluding one of the target samples CE in any of the individual samples (R2 = 0.011, p = 0.085, using four leave-one-out association meta-analyses, to ensure empirical-p = 0.217, β = −0.104 for IMAGE-8; R2 = 0.036, p = 0.188, entirely independent discovery and target samples. PRS were esti- empirical-p = 0.556, β = 0.101 for UCLA; R2 = 0.013, p = 0.407, mated for each target sample using PRSice-2 software (Euesden, empirical-p = 0.761, β = −0.121 for Toronto; R2 = 0.006, p = 0.122, Lewis, & O’Reilly, 2015) (https://www.prsice.info) and applying empirical-p = 0.301, β = 0.083 for Barcelona). standard procedures (imputation quality cut-off using PRSice INFO >0.9, and minor allele frequencies cut-off using PRSice Meta-analysis of all datasets MAF > 0.05) (Choi et al., 2018). PRSice computes PRS by calcu- lating the sum of trait-associated alleles, weighted by the log odds Meta-analysis across all thresholds for RTV showed that the best ratio generated from the discovery GWAS. An R 2 ⩾ 0.1 (250-kb threshold for PRS association with RTV was 0.2 (online window) including all single-nucleotide polymorphisms (SNPs) Supplementary Table S2). At this threshold, the PRS for ADHD ( p 1, p 2 = 1) was used for linkage disequilibrium (LD) clumping was significantly associated with RTV (R 2 = 0.011 p = 0.022, 3154 Isabella Vainieri et al. Fig. 1. Forest plot of the meta-analysis of RTV. The overall estimate from random effects model is represented by the diamond below the individual study estimates. Fig. 2. Forest plot of the meta-analysis of CE. The overall estimate from random effects model is represented by the diamond below the individual study estimates. β = 0.088), with a positive direction. The best threshold for PRS diagnosed with ADHD. Combining our samples in meta-analyses, association with CE was 0.001 (online Supplementary Table S3), our results show that polygenic risk for clinically diagnosed but the association with CE did not reach significance CE ADHD is positively associated with higher RTV, but not with (R 2 = 0.011, p = 0.732, β = 0.013). Heterogeneity tests showed CE as measured by Go/No Go tasks. These data suggest that com- low heterogeneity across studies for both measures (Q = 3.777, mon genetic variation relevant for ADHD influences attention p = 0.436, I 2 = 13.513% for RTV; Q = 1.195, p = 0.754, I 2 = 0% regulation (RTV) but not response inhibition processes (CE) in for CE). Forest plots for each variable are reported in Figs 1 a clinical ADHD sample. Whether the lack of an association and 2. with CE could reflect possible involvement of rare variants not detectable in this analysis or limited power to detect a potentially smaller association, requires further study. Discussion Our results on RTV build on previous evidence from a smaller sample of children with ADHD showing a significant positive This is one of the largest studies investigating the association association between a latent variable of arousal-alertness and between ADHD PRS and cognitive impairments in individuals PRS for ADHD (Nigg et al., 2018). Of note, the association we Psychological Medicine 3155 observed between PRS for ADHD and RTV was mostly consistent generalizability of our findings to non-European populations across all p value thresholds in the meta-analysis, with only slight requires further investigation. Fourth, the use of different tasks fluctuations in results possibly due to low power. Similarly, our to reflect the two constructs of interest at different sites could results on CE are consistent with a previous population-based have introduced heterogeneity in our data; however, we used ran- study and a clinical study showing no association between poly- dom effects in the meta-analyses to account for between-study genic risk for ADHD and other inhibition measures (Martin variation across sites. A further direction for future research is et al., 2015a; Nigg et al., 2018), although a recent study did report to widen the PRS investigation to additional cognitive impair- an association between PRS for ADHD and interference when ments associated with ADHD. measured with the variance of word interference time in the Overall, polygenic risk associated with clinical ADHD diagno- Stroop test (Chang et al., 2020). Previous twin and sibling ana- sis was associated with higher RTV in individuals with clinically lyses have indicated a degree of shared genetic/familial influences diagnosed ADHD. Our results provide molecular genetic evidence on ADHD and response inhibition (Kuntsi et al., 2006b; Kuntsi that attention regulation and ADHD share common genetic fac- et al., 2010). Further evidence from a sibling study suggested in tors. In other words, ADHD common variants not only contrib- fact two familial cognitive impairment factors for ADHD: a larger ute to risk of ADHD diagnosis, but are also a marker of poorer factor (85% of familial variance of ADHD) related to RTV, and a RTV performance in the context of having such a diagnosis. smaller factor (12.5% of familial variance of ADHD) capturing Further investigation, with bigger sample sizes, is needed to rep- CE and omission errors (an overall measure of task accuracy) licate these findings and to further determine the neurobiological (Kuntsi et al., 2010). The findings from the sibling and twin stud- mechanisms underlying this association. Furthermore, it is ies (Kuntsi et al., 2010, 2014) suggested a potential separation, at unknown whether the findings reported here are specific to the genetic level, between attention regulation and response inhib- ADHD or generalize to other disorders where increased RTV is ition processes in their association with ADHD. It is possible that also observed (such as bipolar disorder, schizophrenia and aut- our current analyses detected the larger factor accounted for by ism) (Brotman, Rooney, Skup, Pine, & Leibenluft, 2009; Kaiser RTV in the sibling analyses (Kuntsi et al., 2010) while the smaller et al., 2008; Karalunas, Geurts, Konrad, Bender, & Nigg, 2014). factor (accounting for CE) could not be detected with the current sample size. Future studies should investigate the genetic correl- Supplementary material. The supplementary material for this article can ation between ADHD and RTV or CE across the whole genome be found at https://doi.org/10.1017/S0033291720005218 using LD score regression, when summary statistics from Acknowledgements. Isabella Vainieri is supported by a 3-year Ph.D. stu- GWAS on the appropriate cognitive traits will be available. dentship awarded by the Social, Genetic and Developmental Psychiatry Although PRSs capture the common risk alleles that contrib- Centre, Institute of Psychiatry, Psychology and Neuroscience, King’s College ute to clinically diagnosed ADHD, they do not incorporate con- London. This research has been conducted using PGC ADHD Resources. tributions from other genetic factors, such as copy number We acknowledge the following grant support: UK Medical Research Council variants (CNVs) and single-nucleotide variants (SNVs) that grant G03001896 (to J. Kuntsi), NIH grants R01MH62873 and may underlie the association of ADHD with RTV or CE. R01MH081803 (to S. Faraone), R01MH116037 (to A. Doyle) and Several studies indicate a role for CNVs and SNVs in contributing RO1NS054124 (to S. Loo), Ministerio de Ciencia e Innovacion, Spain, to ADHD risk (Martin, O’Donovan, Thapar, Langley, & Williams, RTI2018-100968-B-100, AGAUR, Generalitat de Catalunya, 2017-SGR-738, 2015b; Satterstrom et al., 2018; Thapar et al., 2016; Williams et al., the Canadian Institute of Health Research MOP-93696 (to R. Schachar), and European Commission H2020 grants 667302 (CoCA), 643051 and 2012; Williams et al., 2010; Yang et al., 2013). CNVs were shown 728018 (to B. Cormand), 847181 (to J. Buitelaar). Over the course of this to be associated with cognitive features in the general population investigation, Marta Ribasés was a recipient of a Miguel de Servet contract such as general cognitive ability (MacLeod et al., 2012), educa- from the Instituto de Salud Carlos III, Spain (CP09/00119 and CPII15/ tional and occupational attainment (Kendall et al., 2017; 00023). The research leading to these results has received funding from the Männik et al., 2015), and other cognitive phenotypes such as Instituto de Salud Carlos III (PI16/01505, PI17/00289, PI18/01788, PI19/ working memory, episodic memory, speed processing, visual 00721 and PI19/01224), and co-financed by the European Regional attention and fluid intelligence (Kendall et al., 2017). Similarly, Development Fund (ERDF), the Agència de Gestió d’Ajuts Universitaris i de SNVs have been implicated in intellectual disability (Satterstrom Recerca-AGAUR (2017SGR1461) and the Health Research and Innovation et al., 2018). Yet, the extent to which CNVs and other genetic var- Strategy Plan (PERIS-SLT006/17/287), Generalitat de Catalunya, Spain. iants may contribute to cognitive impairments in individuals with Finally, we would also like to acknowledge the ADHD Working Group of the Psychiatric Genomics Consortium. Correspondence and requests for mate- ADHD is poorly understood and is an important direction for rials should be addressed to Prof Jonna Kuntsi. future research. Although this is the largest study to date to investigate RTV Conflict of interest. Professor Jonna Kuntsi has given talks at educational and CE with a cutting-edge PRS method in a sample of indivi- events sponsored by Medice; all funds are received by King’s College duals with clinically diagnosed ADHD, certain limitations need London and used for studies of ADHD. Professor Philip Asherson has received to be considered. First, our individual study analyses were under- funding for research by Vifor Pharma and has given sponsored talks and been powered due to the small sample sizes available in each single an advisor for Shire, Janssen-Cilag, Eli-Lilly, Flynn Pharma and Pfizer, regard- study. To increase statistical power, we analyzed the target studies ing the diagnosis and treatment of ADHD. All funds are received by King’s with meta-analyses, reaching a combined sample size of n = 743 College London and used for studies of ADHD. Professor Iris Manor has ADHD participants for RTV and n = 679 ADHD participants received funding for research by Alcobra Ltd., and from Enzymotec Ltd., and all funds were received by Geha MHC for research. She has also given for CE; yet future studies, ideally with larger samples, are needed sponsored talks and been an advisor for Shire, Janssen-Cilag, Teva Israel, to replicate these results. Second, the age range of our participants Medison and Novartis Israel regarding the diagnosis and treatment of was wide (8–45 years old). It would be informative in future larger ADHD. In the last 36 months, Professor Hans-Christoph Steinhausen has studies to explore results separately for participants of different worked as a speaker for an educational event sponsored by Medice and has age groups (children, young adults and older adults). Third, our received book royalties from Cambridge University Press, Elsevier, Hogrefe, study included only participants of European ancestry; the Huber, Klett and Kohlhammer publishers. Professor Banaschewski served 3156 Isabella Vainieri et al. in an advisory or consultancy role for Lundbeck, Medice, Neurim of Genetics, Microbiology, and Statistics, Faculty of Biology, University of Pharmaceuticals, Oberberg GmbH, Shire and Infectopharm. He received con- Barcelona, Catalonia, Spain; 23Department of Experimental Clinical and Health ference support or speaker’s fee by Lilly, Medice and Shire. He received royal- Psychology, Ghent University, Gent, Belgium; 24Department of Child and ties from Hogrefe, Kohlhammer, CIP Medien, Oxford University Press; the Adolescent Psychiatry and Psychotherapy, University Medical Center Göttingen, present work is unrelated to these relationships. Professor Jan K Buitelaar Goettingen, Germany; 25Department of Clinical Neuropsychology, Vrije has been in the past 3 years a consultant to/member of advisory board of/ Universiteit, Amsterdam, The Netherlands; 26Department of Child and and/or speaker for Takeda/Shire, Roche, Medice, Angelini, Janssen and Adolescent Psychiatry, University of Zurich, Zurich, Switzerland; 27Clinical Servier. He is not an employee of any of these companies, and not a stock Psychology and Epidemiology, Institute of Psychology, University of Basel, shareholder of any of these companies. He has no other financial or material Basel, Switzerland; 28Department of Child and Adolescent Psychiatry, University support, including expert testimony, patents, and royalties. In the past year, of Southern Denmark, Odense, Denmark; 29Centre of Child and Adolescent Professor Faraone received income, potential income, travel expenses continu- Mental Health, Capital Region Psychiatry, Copenhagen, Denmark; 30Center for ing education support and/or research support from Takeda, OnDosis, Tris, Genomic Medicine, Massachusetts General Hospital, Boston, MA, USA and 31 Otsuka, Arbor, Ironshore, Rhodes, Akili Interactive Labs, Enzymotec, Massachusetts General Hospital and Harvard Medical School, Boston, Sunovion, Supernus and Genomind. With his institution, he has US patent MA, USA US20130217707 A1 for the use of sodium–hydrogen exchange inhibitors in the treatment of ADHD. He also receives royalties from books published by Guilford Press: Straight Talk about Your Child’s Mental Health, Oxford References University Press: Schizophrenia: The Facts and Elsevier: ADHD: Non- Alemany, S., Ribasés, M., Vilor-Tejedor, N., Bustamante, M., Sánchez-Mora, Pharmacologic Interventions. He is Program Director of www.adhdinadults. C., Bosch, R., … Sunyer, J. (2015). New suggestive genetic loci and bio- com. J. Antoni Ramos-Quiroga was on the speakers’ bureau and/or acted as con- logical pathways for attention function in adult attention-deficit/hyperactiv- sultant for Eli-Lilly, Janssen-Cilag, Novartis, Shire, Takeda, Bial, Shionogui, ity disorder. American Journal of Medical Genetics. Part B, Neuropsychiatric Lundbeck, Almirall, Braingaze, Sincrolab, Medice and Rubió in the last 5 Genetics, 168(6), 459–470. https://doi.org/10.1002/ajmg.b.32341. years. He also received travel awards (air tickets + hotel) for taking part in psy- Andreou, P., Neale, B. M., Chen, W., Christiansen, H., Gabriels, I., Heise, A., chiatric meetings from Janssen-Cilag, Rubió, Shire, Takeda, Shionogui, Bial, … Kuntsi, J. (2007). Reaction time performance in ADHD: Improvement Medice and Eli-Lilly. The Department of Psychiatry chaired by him received under fast-incentive condition and familial effects. Psychological Medicine, unrestricted educational and research support from the following companies 37(12), 1703–1715. https://doi.org/10.1017/S0033291707000815. in the last 5 years: Eli-Lilly, Lundbeck, Janssen-Cilag, Actelion, Shire, Ferrer, Börger, N., & van der Meere, J. (2000). Motor control and state regulation in Oryzon, Roche, Psious and Rubió. Professor Barbara Franke received educa- children with ADHD: A cardiac response study. Biological Psychology, 51 tional speaking fees from Medice. Dr Crosbie has received unrestricted funds (2–3), 247–267. https://doi.org/10.1016/S0301-0511(99)00040-X. from DNAgenotek. All funds were received by the Hospital for Sick Children Brikell, I., Larsson, H., Lu, Y., Pettersson, E., Chen, Q., Kuja-Halkola, R., … for studies related to ADHD and neurodevelopmental disorders. Dr Schachar Martin, J. (2018). The contribution of common genetic risk variants for has consulted for Highland Therapeutics, Purdue Pharma, E Lilly Corp and ADHD to a general factor of childhood psychopathology. Molecular Ehave. The other authors report no conflicts of interest. Psychiatry, 25, 1809–1821. https://doi.org/10.1038/s41380-018-0109-2. Brotman, M. A., Rooney, M. H., Skup, M., Pine, D. S., & Leibenluft, E. (2009). 1 Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, Increased intrasubject variability in response time in youths with bipolar Psychology and Neuroscience, King’s College London, London, UK; 2MRC Centre disorder and at-risk family members. Journal of the American Academy for Neuropsychiatric Genetics & Genomics, School of Medicine, Cardiff of Child and Adolescent Psychiatry, 48(6), 628–635. https://doi.org/10. University, Cardiff, UK; 3Department of Psychiatry & Department of 1097/CHI.0b013e3181a27527. Environmental Medicine, Public Health at the Icahn School of Medicine at Chang, S., Yang, L., Wang, Y., & Faraone, S. V. (2020). Shared polygenic risk for Mount Sinai, New York, NY, USA; 4Department of Child and Adolescent ADHD, executive dysfunction and other psychiatric disorders. Translational Psychiatry, Central Institute of Mental Health and Mannheim Medical Faculty, Psychiatry, 10(1), 182. https://doi.org/10.1038/s41398-020-00872-9. University of Heidelberg, Heidelberg, Germany; 5Karakter Child and Adolescent Cheung, C. H. M., McLoughlin, G., Brandeis, D., Banaschewski, T., Asherson, Psychiatry University Center, Nijmegen, The Netherlands; 6Department of P., & Kuntsi, J. (2017). Neurophysiological correlates of attentional fluctu- Cognitive Neuroscience, Donders Institute for Brain, Cognition and Behavior, ation in attention-deficit/hyperactivity disorder. Brain Topography, 30(3), Radboud University Medical Centre, Nijmegen, The Netherlands; 7Department 320–332. https://doi.org/10.1007/s10548-017-0554-2. of Genetics, Microbiology and Statistics, Faculty of Biology, University of Choi, S. W., Mak, T. S. H., & O’Reilly, P. (2018). Tutorial: a guide to perform- Barcelona, Catalonia, Spain; 8Centro de Investigación Biomédica en Red de ing polygenic risk score analyses. Natureprotocols, 15, 2759–2772. https:// Enfermedades Raras (CIBERER), Instituto de Salud Carlos III, Madrid, Spain; doi.org/10.1101/416545. 9 Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Conners, C. K. (2000). Continuous performance test II: Computer program for Catalonia, Spain; 10Institut de Recerca Sant Joan de Déu (IR-SJD), Esplugues de windows technical guide and software manual. Multi-Health Systems Inc. Llobregat, Catalonia, Spain; 11Psychiatry, Neurosciences and Mental Health, The Demontis, D., Walters, R. K., Martin, J., Mattheisen, M., Als, T. D., Agerbo, E., Hospital for Sick Children, University of Toronto, Toronto, Canada; … Neale, B. M. (2019). Discovery of the first genome-wide significant risk 12 Departments of Psychiatry and Neuroscience and Physiology, SUNY Upstate loci for attention deficit/hyperactivity disorder. Nature Genetics, 51(1), Medical University, Syracuse, New York, USA; 13Departments of Human Genetics 63–75. https://doi.org/10.1038/s41588-018-0269-7. and Psychiatry, Donders Institute for Brain, Cognition and Behaviour, Radboud Du Rietz, E., Coleman, J., Glanville, K., Choi, S. W., O’Reilly, P. F., & Kuntsi, J. University Medical Centre, Nijmegen, The Netherlands; 14Department of (2018). Association of polygenic risk for attention-deficit/hyperactivity dis- Psychiatry and Biobehavioral Sciences, Semel Institute for Neuroscience and order with co-occurring traits and disorders. Biological Psychiatry: Cognitive Human Behavior and David Geffen School of Medicine at UCLA, Los Angeles, CA, Neuroscience and Neuroimaging, 3(7), 635–643. https://doi.org/10.1016/j. USA; 15Department of Developmental and Educational Psychology, University of bpsc.2017.11.013. Valencia, Valencia, Spain; 16Geha Mental Health Center, Petah Tikva, Sackler Euesden, J., Lewis, C. M., & O’Reilly, P. F. (2015). PRSice: Polygenic risk score School of Medicine, Tel-Aviv University, Tel-Aviv, Israel; 17Department of Child software. Bioinformatics (Oxford, England), 31(9), 1466–1468. https://doi. and Adolescent Psychiatry and Psychotherapy, University of Duisburg-Essen, org/10.1093/bioinformatics/btu848. Essen, Germany; 18Department of Psychiatry, Hospital Universitari Vall Hale, T. S., Kane, A. M., Tung, K. L., Kaminsky, O., McGough, J. J., Hanada, d’Hebron, Barcelona, Catalonia, Spain; 19Group of Psychiatry, Mental Health G., & Loo, S. K. (2014). Abnormal parietal brain function in ADHD: and Addictions, Vall d’Hebron Research Institute (VHIR), Barcelona, Catalonia, Replication and extension of previous EEG beta asymmetry findings. Spain; 20Biomedical Network Research Centre on Mental Health (CIBERSAM), Frontiers in Psychiatry, 5, 87. https://doi.org/10.3389/fpsyt.2014.00087. Barcelona, Catalonia, Spain; 21Department of Psychiatry and Forensic Medicine, James, S.-N., Cheung, C. H. M., Rijsdijk, F., Asherson, P., & Kuntsi, J. (2016). Universitat Autònoma de Barcelona, Barcelona, Catalonia, Spain; 22Department Modifiable arousal in attention-deficit/hyperactivity disorder and its Psychological Medicine 3157 etiological association with fluctuating reaction times. Biological Psychiatry: Männik, K., Mägi, R., Macé, A., Cole, B., Guyatt, A. L., Shihab, H. A., … Cognitive Neuroscience and Neuroimaging, 1(6), 539–547. https://doi.org/10. Reymond, A. (2015). Copy number variations and cognitive phenotypes 1016/j.bpsc.2016.06.003. in unselected populations. The Journal of the American Medical Kaiser, S., Roth, A., Rentrop, M., Friederich, H.-C., Bender, S., & Weisbrod, M. Association, 313(20), 2044–2054. https://doi.org/10.1001/jama.2015.4845. (2008). Intra-individual reaction time variability in schizophrenia, depres- Martin, J., Hamshere, M. L., Stergiakouli, E., O’Donovan, M. C., & Thapar, A. sion and borderline personality disorder. Brain and Cognition, 66(1), (2015a). Neurocognitive abilities in the general population and composite 73–82. https://doi.org/10.1016/j.bandc.2007.05.007. genetic risk scores for attention-deficit hyperactivity disorder. Journal of Karalunas, S. L., Geurts, H. M., Konrad, K., Bender, S., & Nigg, J. T. (2014). Child Psychology and Psychiatry, and Allied Disciplines, 56(6), 648–656. Annual research review: Reaction time variability in ADHD and autism https://doi.org/10.1111/jcpp.12336. spectrum disorders: Measurement and mechanisms of a proposed trans- Martin, J., O’Donovan, M. C., Thapar, A., Langley, K., & Williams, N. (2015b). diagnostic phenotype. Journal of Child Psychology and Psychiatry, and The relative contribution of common and rare genetic variants to ADHD. Allied Disciplines, 55(6), 685–710. https://doi.org/10.1111/jcpp.12217. Translational Psychiatry, 5, e506. https://doi.org/10.1038/tp.2015.5. Kendall, K. M., Rees, E., Escott-Price, V., Einon, M., Thomas, R., Hewitt, J., … Müller, U. C., Asherson, P., Banaschewski, T., Buitelaar, J. K., Ebstein, R. P., Kirov, G. (2017). Cognitive performance among carriers of pathogenic copy Eisenberg, J., … Steinhausen, H.-C. (2011a). The impact of study design number variants: Analysis of 152000 UK biobank subjects. Biological and diagnostic approach in a large multi-centre ADHD study. Part 1: Psychiatry, 82(2), 103–110. https://doi.org/10.1016/j.biopsych.2016.08.014. ADHD symptom patterns. BMC Psychiatry, 11, 54. https://doi.org/10. Kofler, M. J., Rapport, M. D., Sarver, D. E., Raiker, J. S., Orban, S. A., 1186/1471-244X-11-54. Friedman, L. M., & Kolomeyer, E. G. (2013). Reaction time variability in Müller, U. C., Asherson, P., Banaschewski, T., Buitelaar, J. K., Ebstein, R. P., ADHD: A meta-analytic review of 319 studies. Clinical Psychology Eisenberg, J., … Steinhausen, H.-C. (2011b). The impact of study design Review, 33(6), 795–811. https://doi.org/10.1016/j.cpr.2013.06.001. and diagnostic approach in a large multi-centre ADHD study: Part 2: Kuntsi, J., Andreou, P., Ma, J., Börger, N. A., & van der Meere, J. J. (2005). Dimensional measures of psychopathology and intelligence. BMC Testing assumptions for endophenotype studies in ADHD: Reliability and Psychiatry, 11, 55. https://doi.org/10.1186/1471-244X-11-55. validity of tasks in a general population sample. BMC Psychiatry, 5, 40. Nigg, J. T., Gustafsson, H. C., Karalunas, S. L., Ryabinin, P., McWeeney, S. K., https://doi.org/10.1186/1471-244X-5-40. Faraone, S. V., … Wilmot, B. (2018). Working memory and vigilance as Kuntsi, J., Frazier-Wood, A. C., Banaschewski, T., Gill, M., Miranda, A., Oades, multivariate endophenotypes related to common genetic risk for attention- R. D., … Rijsdijk, F. (2013). Genetic analysis of reaction time variability: deficit/hyperactivity disorder. Journal of the American Academy of Child Room for improvement? Psychological Medicine, 43(6), 1323–1333. and Adolescent Psychiatry, 57(3), 175–182. https://doi.org/10.1016/j.jaac. https://doi.org/10.1017/S0033291712002061. 2017.12.013. Kuntsi, J., Neale, B. M., Chen, W., Faraone, S. V., & Asherson, P. (2006a). The Riglin, L., Eyre, O., Cooper, M., Collishaw, S., Martin, J., Langley, K., … Thapar, IMAGE project: Methodological issues for the molecular genetic analysis of A. (2017). Investigating the genetic underpinnings of early-life irritability. ADHD. Behavioral and Brain Functions, 2, 27. https://doi.org/10.1186/ Translational Psychiatry, 7(9), e1241. https://doi.org/10.1038/tp.2017.212. 1744-9081-2-27. Sánchez-Mora, C., Ramos-Quiroga, J. A., Bosch, R., Corrales, M., Garcia-Martínez, Kuntsi, J., Pinto, R., Price, T. S., van der Meere, J. J., Frazier-Wood, A. C., & I., Nogueira, M., … Ribasés, M. (2015). Case-control genome-wide association Asherson, P. (2014). The separation of ADHD inattention and study of persistent attention-deficit hyperactivity disorder identifies FBXO33 as hyperactivity-impulsivity symptoms: Pathways from genetic effects to cog- a novel susceptibility gene for the disorder. Neuropsychopharmacology, 40(4), nitive impairments and symptoms. Journal of Abnormal Child Psychology, 915–926. https://doi.org/10.1038/npp.2014.267. 42(1), 127–136. https://doi.org/10.1007/s10802-013-9771-7. Satterstrom, F. K., Walters, R. K., Singh, T., Wigdor, E. M., Lescai, F., Kuntsi, J., Rogers, H., Swinard, G., Börger, N., van der Meere, J., Rijsdijk, F., & Demontis, D., … Daly, M. J. (2018). Autism spectrum disorder and atten- Ash-erson, P. (2006b). Reaction time, inhibition, working memory and tion deficit hyperactivity disorder have a similar burden of rare protein- ‘delay aversion’ performance: Genetic influences and their interpretation. truncating variants. Nature Neuroscience, 22(12), 1961–1965. https:// Psychological Medicine, 36(11), 1613–1624. https://doi.org/10.1017/ doi.org/10.1038/s41593-019-0527-8. S0033291706008580. Schachar, R., Logan, G. D., Robaey, P., Chen, S., Ickowicz, A., & Barr, C. Kuntsi, J., Wood, A. C., Rijsdijk, F., Johnson, K. A., Andreou, P., Albrecht, B., (2007). Restraint and cancellation: Multiple inhibition deficits in attention … Asherson, P. (2010). Separation of cognitive impairments in attention- deficit hyperactivity disorder. Journal of Abnormal Child Psychology, 35 deficit/hyperactivity disorder into 2 familial factors. Archives of General (2), 229–238. https://doi.org/10.1007/s10802-006-9075-2. Psychiatry, 67(11), 1159–1167. https://doi.org/10.1001/archgenpsychiatry. Stergiakouli, E., Martin, J., Hamshere, M. L., Heron, J., St Pourcain, B., 2010.139. Timpson, N. J., … Davey Smith, G. (2017). Association between polygenic Kuntsi, J., Wood, A. C., Van Der Meere, J., & Asherson, P. (2009). Why cognitive risk scores for attention-deficit hyperactivity disorder and educational and performance in ADHD may not reveal true potential: Findings from a large cognitive outcomes in the general population. International Journal of population-based sample. Journal of the International Neuropsychological Epidemiology, 46(2), 421–428. https://doi.org/10.1093/ije/dyw216. Society, 15(4), 570–579. https://doi.org/10.1017/S135561770909081X. Swanson, J. M., Schuck, S., Porter, M. M., Carlson, C., Hartman, C. A., Lionel, A. C., Crosbie, J., Barbosa, N., Goodale, T., Thiruvahindrapuram, B., Sergeant, J. A., … Wigal, T. (2012). Categorical and dimensional definitions Rickaby, J., … Scherer, S. W. (2011). Rare copy number variation discovery and evaluations of symptoms of ADHD: History of the SNAP and the and cross-disorder comparisons identify risk genes for ADHD. Science SWAN rating scales. The International Journal of Educational and Translational Medicine, 3(95), 95ra75. https://doi.org/10.1126/scitranslmed. Psychological Assessment, 10(1), 51–70. https://www.ncbi.nlm.nih.gov/ 3002464. pubmed/26504617. Logan, G. D., Cowan, W. B., & Davis, K. A. (1984). On the ability to inhibit Thapar, A., Martin, J., Mick, E., Arias Vásquez, A., Langley, K., Scherer, S. W., simple and choice reaction time responses: A model and a method. … Holmans, P. (2016). Psychiatric gene discoveries shape evidence on Journal of Experimental Psychology. Human Perception and Performance, ADHD’s biology. Molecular Psychiatry, 21(9), 1202–1207. https://doi.org/ 10(2), 276–291. https://doi.org/10.1037//0096-1523.10.2.276. 10.1038/mp.2015.163. Loo, S. K., Hale, T. S., Macion, J., Hanada, G., McGough, J. J., McCracken, J. T., Uebel, H., Albrecht, B., Asherson, P., Börger, N. A., Butler, L., Chen, W., … & Smalley, S. L. (2009). Cortical activity patterns in ADHD during arousal, Banaschewski, T. (2010). Performance variability, impulsivity errors and activation and sustained attention. Neuropsychologia, 47(10), 2114–2119. the impact of incentives as gender-independent endophenotypes for https://doi.org/10.1016/j.neuropsychologia.2009.04.013. ADHD. Journal of Child Psychology and Psychiatry, and Allied Disciplines, MacLeod, A. K., Davies, G., Payton, A., Tenesa, A., Harris, S. E., Liewald, D. … 51(2), 210–218. https://doi.org/10.1111/j.1469-7610.2009.02139.x. Deary, I. J. (2012). Genetic copy number variation and general cognitive abil- van der Meere, J., Stemerdink, N., & Gunning, B. (1995). Effects of presenta- ity. PLoS ONE, 7(12), e37385. https://doi.org/10.1371/journal.pone.0037385. tion rate of stimuli on response inhibition in ADHD children with and 3158 Isabella Vainieri et al. without tics. Perceptual and Motor Skills, 81(1), 259–262. https://doi.org/10. attention deficit hyperactivity disorder: The role of rare variants and dupli- 2466/pms.1995.81.1.259. cations at 15q13.3. The American Journal of Psychiatry, 169(2), 195–204. van Rooij, D., Hartman, C. A., Mennes, M., Oosterlaan, J., Franke, B., https://doi.org/10.1176/appi.ajp.2011.11060822. Rom-melse, N., … Hoekstra, P. J. (2015). Altered neural connectivity during Williams, N. M., Zaharieva, I., Martin, A., Langley, K., Mantripragada, K., response inhibition in adolescents with attention-deficit/hyperactivity dis- Fossdal, R., … Thapar, A. (2010). Rare chromosomal deletions and dupli- order and their unaffected siblings. NeuroImage. Clinical, 7, 325–335. cations in attention-deficit hyperactivity disorder: A genome-wide analysis. https://doi.org/10.1016/j.nicl.2015.01.004. The Lancet, 376(9750), 1401–1408. https://doi.org/10.1016/S0140-6736(10) Vuijk, P. J., Martin, J., Braaten, E. B., Genovese, G., Capawana, M. R., O’Keefe, S. 61109-9. M., … Doyle, A. E. (2019). Translating discoveries in attention-deficit/hyper- Yang, L., Neale, B. M., Liu, L., Lee, S. H., Wray, N. R., Ji, N., … Subgroup, activity disorder genomics to an outpatient child and adolescent psychiatric P. G. C. A. (2013). Polygenic transmission and complex neuro developmen- cohort. Journal of the American Academy of Child and Adolescent tal network for attention deficit hyperactivity disorder: Genome-wide Psychiatry, 59(8), 964–977. https://doi.org/10.1016/j.jaac.2019.08.004. association study of both common and rare variants. American Journal of Williams, N. M., Franke, B., Mick, E., Anney, R. J. L., Freitag, C. M., Gill, M., Medical Genetics. Part B, Neuropsychiatric Genetics, 162B(5), 419–430. … Faraone, S. V. (2012). Genome-wide analysis of copy number variants in https://doi.org/10.1002/ajmg.b.32169.