Research Topic(s): This study investigates cerebellar-related sensorimotor functions, visuospatial perception, and visuospatial abilities in adults with ADHD and ASD. We, evaluated cognitive and motor skills using a comprehensive set of tests in patients with medicated ADHD, unmedicated ADHD, ASD, comorbid ADHD and ASD, and a control group.
Main Findings: All patient groups exhibited significantly worse performances than controls in sensorimotor functions, visuospatial perception, and abilities. We observed significant associations between sensorimotor functions and visuospatial perception and abilities. The findings revealed clear deficits in all patient groups, supporting the theory that ADHD and ASD are linked to cerebellar dysfunction.
Study Contributions: This work identify deficits in sensorimotor functions, visuospatial perception, and visuospatial abilities in adults with a neuropsychiatric diagnosis. The majority of patients (over 60% in each group) had a mild disorder severity index. New cognitive and motor tests emerged from this study, strengthening clinical evaluations through objective assessment tools.
Introduction
The number of individuals with neuropsychiatric diagnoses is increasing. More than 366 million individuals worldwide have attention deficit hyperactivity disorder (ADHD) as of 2020 (Song et al., Reference Song, Zha, Yang, Zhang, Li and Rudan2021). Approximately, 120 million individuals worldwide have autism spectrum disorder (ASD) (Talantseva et al., Reference Talantseva, Romanova, Shurdova, Dolgorukova, Sologub, Titova, Kleeva and Grigorenko2023).
Correctly diagnosing neuropsychiatric disorders (NPD) is challenging, in part because there are no biomarkers that can identify ADHD and ASD with high accuracy (Bonvicini et al., Reference Bonvicini, Cortese, Maj, Baune, Faraone and Scassellati2020; Emser et al., Reference Emser, Johnston, Steele, Kooij, Thorell and Christiansen2018; Jensen et al., Reference Jensen, Lane, Werner, McLees, Fletcher and Frye2022). It is important to develop tools with which we can identify and, if possible, differentiate ADHD from ASD because pharmacological treatments are available only for ADHD.
Adults with ADHD and ASD show neuropsychological deficits in multiple cognitive domains (Barkley, Reference Barkley1997; Cundari et al., Reference Cundari, Vestberg, Gustafsson, Gorcenco and Rasmussen2023; Goldberg et al., Reference Goldberg, Mostofsky, Cutting, Mahone, Astor, Denckla and Landa2005; Hlavatá et al., Reference Hlavatá, Kašpárek, Linhartová, Ošlejšková and Bareš2018; Russo et al., Reference Russo, Flanagan, Iarocci, Berringer, Zelazo and Burack2007; Wilson et al., Reference Wilson, Happé, Wheelwright, Ecker, Lombardo, Johnston, Daly, Murphy, Spain, Lai, Chakrabarti, Sauter and Murphy2014). Deficits in executive functions are a key feature of ADHD (Barkley, Reference Barkley1997; Mohamed et al., Reference Mohamed, Butzbach, Fuermaier, Weisbrod, Aschenbrenner, Tucha, Tucha and Garcia2021). Neurobiological models have focused on the attentional network consisting of the prefrontal cortex (PFC), anterior cingulate, and basal ganglia (Martella et al., Reference Martella, Aldunate, Fuentes and Sánchez-Pérez2020). The functional connectivity between the cerebellum and PFC was altered in ASD (Mittleman et al., Reference Mittleman, Goldowitz, Heck and Blaha2008). The main deficits associated with ASD concern sensory processing, and social communication. However, executive function, motor timing, and visual and verbal episodic memory are also compromised (Hlavatá et al., Reference Hlavatá, Kašpárek, Linhartová, Ošlejšková and Bareš2018; Maw et al., Reference Maw, Beattie and Burns2024; Russo et al., Reference Russo, Flanagan, Iarocci, Berringer, Zelazo and Burack2007).
We know less about sensorimotor functions, visuospatial perception, and visuospatial abilities in these disorders (Aycicegi-Dinn et al., Reference Aycicegi-Dinn, Dervent-Ozbek, Yazgan, Bicer and Dinn2011; Braconnier & Siper, Reference Braconnier and Siper2021; Callahan et al., Reference Callahan, Ramakrishnan, Shammi, Bierstone, Taylor, Ozzoude, Goubran, Stuss and Black2022; Minshew et al., Reference Minshew, Goldstein and Siegel1997; Schreiber et al., Reference Schreiber, Javorsky, Robinson and Stern1999; Wilson et al., Reference Wilson, Happé, Wheelwright, Ecker, Lombardo, Johnston, Daly, Murphy, Spain, Lai, Chakrabarti, Sauter and Murphy2014). Some studies have linked impairments in sensorimotor functions with ADHD and ASD (Braconnier & Siper, Reference Braconnier and Siper2021; Guo et al., Reference Guo, Fuermaier, Koerts, Mueller, Diers, Mroß, Mette, Tucha and Tucha2021; Marinopoulou et al., Reference Marinopoulou, Lugnegård, Hallerbäck, Gillberg and Billstedt2016; Nikolas et al., Reference Nikolas, Marshall and Hoelzle2019; Tinius, Reference Tinius2003; Wilson et al., Reference Wilson, Happé, Wheelwright, Ecker, Lombardo, Johnston, Daly, Murphy, Spain, Lai, Chakrabarti, Sauter and Murphy2014). Similarly, there is evidence suggesting that visuospatial perception and visuospatial abilities are compromised, but no previous studies included adults with different clinical neuropsychiatric categories (Aycicegi-Dinn et al., Reference Aycicegi-Dinn, Dervent-Ozbek, Yazgan, Bicer and Dinn2011; Braconnier & Siper, Reference Braconnier and Siper2021; Callahan et al., Reference Callahan, Ramakrishnan, Shammi, Bierstone, Taylor, Ozzoude, Goubran, Stuss and Black2022; Minshew et al., Reference Minshew, Goldstein and Siegel1997; Schreiber et al., Reference Schreiber, Javorsky, Robinson and Stern1999; Wilson et al., Reference Wilson, Happé, Wheelwright, Ecker, Lombardo, Johnston, Daly, Murphy, Spain, Lai, Chakrabarti, Sauter and Murphy2014). One reason for this is that the previous Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) had specified that an ASD diagnosis was an exclusion criterion for ADHD, thereby research was limited to this clinical comorbidity (Leitner, Reference Leitner2014).
ADHD and ASD are associated with differences in cerebellar morphology and function (Duan et al., Reference Duan, Jiang, Rootes-Murdy, Schoenmacker, Arias-Vasquez, Buitelaar, Hoogman, Oosterlaan, Hoekstra, Heslenfeld, Hartman, Calhoun, Turner and Liu2021; Riva et al., Reference Riva, Annunziata, Contarino, Erbetta, Aquino and Bulgheroni2013). The cerebellum and cerebrum are ∼ 4% smaller in ADHD patients compared to controls (Castellanos et al., Reference Castellanos, Lee, Sharp, Jeffries, Greenstein, Clasen, Blumenthal, James, Ebens, Walter, Zijdenbos, Evans, Giedd and Rapoport2002; Vaidya, Reference Vaidya2011). Similarly, a post-mortem study demonstrated that ASD patients had abnormal cerebellar morphology (Amaral et al., Reference Amaral, Schumann and Nordahl2008). Morphological abnormalities in different cerebellar subregions produce distinct behavioral symptoms related to the functional disruption of specific cerebro-cerebellar circuits (Silk et al., Reference Silk, Vance, Rinehart, Bradshaw and Cunnington2009; Stoodley, Reference Stoodley2014, Reference Stoodley2016), which can explain why cerebellar abnormalities can lead to ADHD and ASD.
The cerebellum plays a major role in motor control and sensorimotor integration, but also in visual search, spatial mapping, visuospatial perception and visuospatial ability (Cundari et al., Reference Cundari, Vestberg, Gustafsson, Gorcenco and Rasmussen2023; King et al., Reference King, Hernandez-Castillo, Poldrack, Ivry and Diedrichsen2019; Kuhn et al., Reference Kuhn, Gullett, Boutzoukas, Bohsali, Mareci, FitzGerald, Carney and Bauer2018). More specifically, the cerebellum is involved in spatial representation and receives input from motor and vestibular systems to acquire information about self-motion (Ango & Dos Reis, Reference Ango and Dos Reis2019; Manto et al., Reference Manto, Bower, Conforto, Delgado-García, da Guarda, Gerwig, Habas, Hagura, Ivry, Mariën, Molinari, Naito, Nowak, Oulad Ben Taib, Pelisson, Tesche, Tilikete and Timmann2012; Paulin, Reference Paulin1993; Yoo & Mihaila, Reference Yoo and Mihaila2022).
The cerebellum interacts with both the dorsal and ventral stream. Moreover, through projections to hippocampal place cells, it offers updates on an individual location within the environment, based on the existing hippocampal encoding (Baumann et al., Reference Baumann, Borra, Bower, Cullen, Habas, Ivry, Leggio, Mattingley, Molinari, Moulton, Paulin, Pavlova, Schmahmann and Sokolov2015; Chinellato & del Pobil, Reference Chinellato and del Pobil2016; Moser et al., Reference Moser, Kropff and Moser2008; Ramnani et al., Reference Ramnani, Toni, Passingham and Haggard2001; Rochefort et al., Reference Rochefort, Arabo, André, Poucet, Save and Rondi-Reig2011, Reference Rochefort, Lefort and Rondi-Reig2013). This allows for a mechanism whereby self-motion information is integrated into the cognitive map to allow for coordinated movement in space (Rochefort et al., Reference Rochefort, Lefort and Rondi-Reig2013).
Moreover, the cerebellum plays a crucial role in processing self-motion information for spatial navigation, constructing a spatial representation in the hippocampus, and guiding optimal trajectories toward goals (Rochefort et al., Reference Rochefort, Lefort and Rondi-Reig2013). Additionally, deficits in cerebellar function have been associated with impairments in visuospatial abilities, particularly in complex and mentally demanding visual tasks (Molinari & Leggio, Reference Molinari and Leggio2007; Molinari et al., Reference Molinari, Petrosini, Misciagna and Leggio2004). Evidence suggests that cerebellar regions involved in eye movements also contribute to visual guidance and visuospatial attention (Striemer et al., Reference Striemer, Chouinard, Goodale and de Ribaupierre2015). Coping with inherent sensory feedback delays during motor activities, the central nervous system must accurately estimate future motor states to ensure precise movements. Considering all the available evidence, it is clear that the cerebellum plays a role in sensorimotor and visuospatial functions (King et al., Reference King, Hernandez-Castillo, Poldrack, Ivry and Diedrichsen2019; Kuhn et al., Reference Kuhn, Gullett, Boutzoukas, Bohsali, Mareci, FitzGerald, Carney and Bauer2018).
This study aims to measure sensorimotor functions, visuospatial perception, and visuospatial abilities in adults with ADHD and ASD, and to test whether there are clinically relevant subgroup differences between four clinical neuropsychiatric groups. We hypothesize that these functions are lower in patients with ADHD and ASD than in controls. Furthermore, we want to study how sensorimotor functions can affect visuospatial perception and visuospatial abilities. This study also aims to examine the impact of central nervous system (CNS) stimulants on cognitive functions in adults with ADHD.
This is the first study to test these functions in adults belonging to four different clinical groups, which correspond to the current neuropsychiatric population. Previous research has shown structural and functional cerebellar alterations in individuals with ADHD and ASD, suggesting shared neurobiological factors (Cundari et al., Reference Cundari, Vestberg, Gustafsson, Gorcenco and Rasmussen2023; Stoodley, Reference Stoodley2014, Reference Stoodley2016). New Swedish guidelines emphasize a more holistic approach when assessing these disorders – resulting in a large group of patients with combined ADHD and ASD that we know little about (Swedish national guidelines: Adhd and autism, 2024).
Material and methods
Study design
This was a cross-sectional observational study. The primary variables of interest were outcomes on tests measuring sensorimotor functions, visuospatial perception, visuospatial abilities, diagnoses, and medication status. We also examined potential moderating variables including demographic variables, anxiety and depression symptoms, and insomnia scores. Before conducting any tests, the participants signed an informed consent form. The participants went through a clinical interview to assess their mental and physical health, past health issues, sleep difficulties, language disorders, hereditary conditions, alcohol and drug use, and current medications. In addition, participants were asked about their dominant hand, native language, and their level of education. The human data included in this manuscript were obtained in compliance with the Helsinki Declaration (World Medical Association, 2001).
Description of the participants
A total of 272 adults, 188 patients and 84 controls, were tested in 2023–2024 (see Table 1). The patients were recruited from the Unit of Adult Neuropsychiatry from Helsingborg´s hospital, Sweden. We included patients with unmedicated ADHD (U-ADHD), medicated ADHD (M-ADHD), ASD, and the combination of unmedicated ADHD and ASD (COMBO). The ADHD and ASD severity index was assessed based on DSM-V criteria (1–Mild, 2–Moderate, or 3–Severe). The inclusion criteria were: (1) age between 18 and 75, (2) ability to speak Swedish, and (3) a diagnosis of ADHD and/or ASD documented in their medical journal. Exclusion criteria were: previous brain injury, intellectual disability, and continuous substance abuse. We excluded 2 patients with undiagnosed intellectual disability, detected during neurocognitive assessment; 6 patients who showed neuropsychiatric symptomatology but did not fulfill diagnostic criteria, 1 patient with cardiovascular disease, 1 patient with ASD medicated with methylphenidate, 23 patients medicated for ADHD and who also had an ASD diagnosis. The final sample consisted of 155 patients. The distribution of patients in the different groups is shown in Table 1. Details of comorbidities and medication dosages are provided in the supplementary materials. For the controls, the inclusion criteria were: (1) age between 18–75, and (2) ability to speak Swedish. Controls were excluded if they had a psychiatric diagnosis, neurological disorder, intellectual disability, previous brain injury, sleep disorders, language impairments or continuous substance abuse. Of the 84 controls, 78 fulfilled the inclusion criteria.
Table 1. Demographic and clinical characteristics
Note: The means, standard deviations, and ranges (minimum and maximum values) are reported. All p-values were determined using the Kruskal–Wallis test. U-ADHD = unmedicated ADHD, M-ADHD = medicated ADHD, ASD = autism spectrum disorder, COMBO = both unmedicated ADHD and ASD. The severity indices for COMBO (1, 2, and 3) were calculated according to the severity indices for ADHD and ASD as defined in the DSM-V. n.a.= indicates not applicable and n = the sample size for each group.
Neurocognitive assessments for sensorimotor functions and processing speed
Coding and symbol search, subtests of Wechsler Adult Intelligence Scale (WAIS-IV)
Both subtests measure sensorimotor integration, visual scanning, and processing speed. The coding test primarily measures incidental learning and visual-motor coordination whereas the symbol search test primarily measures visual discrimination and visual scanning (Lichtenberger & Kaufman, Reference Lichtenberger and Kaufman2012; Wechsler, Reference Wechsler2008). The variables selected for the analysis were the total raw score.
Motor speed, trail making test of Delis-Kaplan Executive Function System (D-KEFS)
This subtest measures motor speed without cognitive load. It provides a direct, isolated measure of the motor demands. The variable selected for the analysis was total time, up to a maximum of 150 s (delis et al., Reference Delis, Kaplan and Kramer2001).
Motor and perception test, Bender-Gestalt II
These supplemental tests measure fine motor and perceptual abilities. A correct response is scored as one (1) and incorrect response is scored zero (0) (Brannigan & Decker, Reference Brannigan and Decker2006). As an explorative measure, we also recorded the time it took to complete the task in seconds. The variable selected for the analysis was total time. We calculated the concurrent validity for the exploratory neuropsychological tests. Detailed information regarding these analyses can be found in the supplementary materials.
Motor screening task and Reaction time, Cambridge Neuropsychological Test Automated Battery (CANTAB)
All CANTAB tasks were administered on an iPad (9rh generation; 10.5-inch screen). Motor Screening Task (MOT) measures sensorimotor functions and processing speed. Outcome measures were speed of response and accuracy of pointing. It is used as a baseline measure for motor abilities (Boyle et al., Reference Boyle, Northoff, Barbey, Fregni, Jahanshahi, Pascual-Leone and Sahakian2023). The variable selected for the analysis was MOTML (the mean latency from the display of a stimulus to a correct response to that stimulus). Reaction Time (RTI) measures processing, psychomotor speed, and movement (CANTAB®[Cognitive assessment software], 2019). Measures include reaction time and movement time for both the simple and five-choice modes. The variables selected for analysis were RTIFMRT (the mean time taken for a subject to release the response button after the presentation of a target stimulus), and RTIFMMT (the mean time taken for a subject to select the target stimulus after releasing the response button). These variables are measured in milliseconds.
Cognitive assessments for visuospatial abilities and visuospatial perception
Cube analysis and silhouettes, subtests of Visual Object and Space Perception Battery (VOSP)
Cube analysis measures visuospatial perception and visual discrimination. silhouettes measure visuospatial perception and the ability to recognize common objects and animals depicted from unconventional perspectives (Warrington & James, Reference Warrington and James1991). The variables selected for the analysis were total raw scores. We calculated the concurrent validity for the exploratory neuropsychological tests. Detailed information regarding these analyses can be found in the supplementary materials.
Copy task, Visual Reproduction II (VR-II) of Wechsler Memory Scale III (WMS-III)
The participants were instructed to copy five figures. The figures were presented one at a time. Each stimulus was available to the participant continuously, thus, removing the memory aspect and focusing more on visuospatial abilities and visuomotor integration (Tulsky et al., Reference Tulsky, Saklofske, Chelune, Heaton, Ivnik, Bornstein, Prifitera and Ledbetter2003). We recorded the time it took to copy each figure. Participants were instructed to quickly and accurately reproduce the figure and inform the test administrator when finished. Scoring of the five figures was done in accordance with the WMS-III manual (Tulsky, Reference Tulsky2004). The criteria measure accuracy and spatial positioning. The scoring range was: 0–10 for figure 1; 0–10 for figure 2; 0–18 for figure 3; 0–34 for figure 4; 0–32 for figure 5; 0–104 in total. The variables selected for analysis were total raw scores and total time in seconds, as experimental variables. We calculated the concurrent validity for the exploratory neuropsychological tests. Detailed information regarding these analyses can be found in the supplementary materials.
Cerebellar tests
Finger tapping
This test measures sensorimotor synchronization. It is one of the many tools available to assess motor control and the integrity of the neuromuscular system (Gao et al., Reference Gao, Mei and Chen2015; Gustafsson et al., Reference Gustafsson, Kjell, Cundari, Larsson, Edbladh, Madison, Kazakova and Rasmussen2023). We evaluated isochronous serial interval production (ISIP) tasks (Madison, Reference Madison2001, Reference Madison2006), that has been linked to cerebellar function (Rivkin et al., Reference Rivkin, Vajapeyam, Hutton, Weiler, Hall, Wolraich, Yoo, Mulkern, Forbes, Wolff and Waber2003; Turesky et al., Reference Turesky, Olulade, Luetje and Eden2018). The participant must synchronize taps on the space bar to a rhythmic sound (synchronization phase) and continue pressing the space bar at the same pace when the rhythmic sound stops (production phase). This procedure was repeated five times. Each repetition included 15 paced intervals followed by 70 self-paced intervals. The inter-onset interval (IOI) was 524 ms. The mean inter-tap interval and two measures of variance (drift and local) were used for the ISIP data. Response IOIs shorter than 400 ms and longer than 650 ms were excluded from the analyses. The variables selected for the analysis were production mean (ms), local (ms), and drift (ms).
Prism adaptation
This test measures sensorimotor coordination following changes in visual input and upper limb coordination. Previous studies demonstrate a relationship between performance in prism adaptation tasks and cerebellar function (Küper et al., Reference Küper, Wünnemann, Thürling, Stefanescu, Maderwald, Elles, Göricke, Ladd and Timmann2014; Pisella et al., Reference Pisella, Rossetti, Michel, Rode, Boisson, Pélisson and Tilikete2005). Participants stood at arm’s length from a measuring tape attached to the wall. On each trial, the participant was instructed to look at the measuring tape and identify the midpoint of the measuring tape, then close their eyes and point to the target. While still pointing and standing in the same place, the participant opened their eyes to see where he/she had pointed. This procedure was repeated 10 times. For the next 10 trials, the participants wore prism glasses that displaced the visual field laterally by 15 degrees. Wearing the prisms usually results in a pointing error that gradually decreases as the participant adapts to the new visuomotor relationship. In the last ten trials (21–30), the glasses were taken off again. Removing the glasses typically results in a pointing error in the direction opposite to the error that occurred when putting the prisms on – often to the surprise of the participant. On each trial, the test administrator wrote down the error in centimeters. A positive value indicates that the participant pointed to the left of the target, and a negative value indicates that the participant pointed to the right of the target. For the analysis we selected the average errors on the first trial after putting the glasses on (trial 11), the average of first trial after taking the glasses off (trial 21) like in the previous study (Gustafsson et al., Reference Gustafsson, Kjell, Cundari, Larsson, Edbladh, Madison, Kazakova and Rasmussen2023) and we create a new variable, the average of errors of all 30 trials.
Self-report scales
Hospital Anxiety and Depression Scale (HADS)
The HADS is a self-report questionnaire that contains 14 items, 7 items measuring cognitive and emotional aspects of depression, specifically anhedonia, and the remaining 7 items focusing on cognitive and emotional aspects of anxiety (Smarr & Keefer, Reference Smarr and Keefer2011; Zigmond & Snaith, Reference Zigmond and Snaith1983). High scores indicate greater severity. Both HADS-A (α=0.87) and HADS-D (α=0.81) have high reliability.
Insomnia Severity Index (ISI)
The ISI is a 7-item self-report questionnaire assessing insomnia severity and impact in the last two weeks (Bastien et al., Reference Bastien, Vallières and Morin2001). High scores indicate greater severity. ISI has high reliability (α=0.79)
Statistical analysis
Statistical analyses were conducted using IBM SPSS Statistics 28.0. Results are expressed as frequencies (percentages) or mean ± SD. To compare quantitative variables across five groups, we used analysis of variance (ANOVA). Outliers (beyond±3 SD) were excluded to ensure that no samples exhibited skewness (beyond –3 and + 3) or kurtosis (beyond –2 and + 2). Normality was evaluated using the Shapiro–Wilk test, and the Levene test was utilized to assess the equality of variances. When the assumptions for ANOVA were not fulfilled, the Kruskal–Wallis test was applied. Residuals were checked using the Quick plot function in SPSS. To assess differences between groups we applied corrections for multiple comparisons, including adjusting the alpha level with the Bonferroni correction to control for the risk of false positives. In the results we only reported differences that were significant after Bonferroni correction. A two-way two-tailed ANOVA was used to evaluate the main and interaction effects of total time (speed) and raw scores on the visuospatial copy task. The Pearson or Spearman rank correlation tests were used to analyze the validity of the exploratory tests. We calculated the split-half reliability of the exploratory tests, and we reported the Spearman–Brown coefficients. Receiver Operating Characterisitc (ROC) analyses were performed to investigate the sensitivity and specificity of the tests in predicting neuropsychiatric diagnoses. In Figure 1, we reported only the tests with an AUC >0.80 (indicating excellent discrimination, 0.80–0.90). Exact p-values are reported in the tables, with statistical significance set at p < 0.05. Effect sizes are reported using eta squared (η2). The effect sizes are classified as small (>0.01), medium (>0.06), or large (>0.14) (Lakens, Reference Lakens2013). Linear and multiple regression analyses were conducted to control for the effects of education level, sleep, anxiety, and depression on neurocognitive performances, as well as to investigate the potential negative impact of sensorimotor deficits on visual perception and visuospatial abilities. Controlling for sleep, anxiety, and depression ensures that research findings on neurocognitive outcomes are accurate and not influenced by these factors (Brownlow et al., Reference Brownlow, Miller and Gehrman2020; Dorenkamp et al., Reference Dorenkamp, Irrgang and Vik2023; Kriesche et al., Reference Kriesche, Woll, Tschentscher, Engel and Karch2023).
Figure 1. ROC Curves for key variables among all neuropsychiatric disorder groups.
Note. The graph displays the specificity and sensitivity in predicting correct diagnosis among all the groups in the study. The selected tests have area under the curve (AUC) > 0.80. In the graphs we report individual AUC for each test variable for all subclinical group. A) ROC Curves for unmedicated attention deficit hyperactivity disorder (ADHD), B) ROC Curves for medicated ADHD, C) ROC Curves for autism spectrum disorder (ASD), D) ROC Curves for the combination of unmedicated ADHD and ASD.
Results
Participants
Table 1 provides an overview of demographics and clinical variables, as well as their distributions. A Kruskal–Wallis analysis showed that there was significant differences in age and years of education when comparing all groups. However, the post hoc analysis did not reveal any significant differences when comparing specific groups. There were significant differences in anxiety, depression, and sleep scores between the control and NPD groups, but post hoc analyses did not reveal any differences between the different clinical subgroups. The severity distribution shows that most of the patients in each of the NPD groups exhibit a mild severity index (60.6–83.9%) evaluated according to the DSM-V. The percentage of patients with moderate severity ranged from 23.1–32.7% in the different NPD groups. In the M-ADHD group, 15.4% of the patients had the highest severity index (see Table 1 for details).
Sensorimotor functions
Table 2 provides an overview of the test performances of sensorimotor functions. To assess whether NPD groups differ from controls on sensorimotor functions, we examined eight variables: (1) total raw scores of Coding and (2) Symbol Search; (3) total time of Motor Speed, (4) Motor Time, (5) Perception Time; (6) MOTML of MOT, (7) RTIFMRT, (8) and RTIFMTT of Reaction Time.
Table 2. Descriptive statistics for the five groups on test performances of sensorimotor functions
Note: The mean, standard deviation, minimum, and maximum values, effect sizes, F-statistic values for ANOVA, and H-statistic values for the Kruskal–Wallis test are reported. U-ADHD = unmedicated ADHD, M-ADHD = medicated ADHD, ASD = autism spectrum disorder, COMBO = both unmedicated ADHD and ASD. Sample sizes vary across groups due to the exclusion of outliers. “KW” indicates p-values calculated using the Kruskal–Wallis test, while ANOVA was used for all other variables. Asterisks (*) denote H-statistic values for the Kruskal–Wallis test. rs = raw scores, sec = seconds, ms = milliseconds, and n = the sample size for each group. MOTML = the mean latency from the display of a stimulus to a correct response to that stimulus, RTIFMRT = the mean time taken for a subject to release the response button after the presentation of a target stimulus, RTIFMMT = the mean time taken for a subject to select the target stimulus after releasing the response button.
As reported in table 2, the results showed significant differences in performance with large effect size on Perception Time, Coding, Symbol Search, Motor Speed, with medium effect size on RTIFMMT of Reaction Time, Motor Time, and MOTML of MOT, with small effect size on RTIFMRT of Reaction Time.
We found after Bonferroni correction significant differences between controls and all four neuropsychiatric groups for Coding, Symbol Search, Motor Speed, Motor Time and Perception Time. Additionally, we found significant differences only in the MOTML scores of the MOT between controls and the ASD group (p< .001), as well as between the M-ADHD and ASD groups (p= .016). RTIFMRT of Reaction Time shows significant differences only between controls and U-ADHD (p = .029). Significant differences emerged in the RTIFMMT of Reaction Time, between controls and U-ADHD (p = .008), controls and ASD (p < .001), controls and COMBO (p = .013), ASD and M-ADHD (p = .003). No other significant differences emerged between the remaining clinical subgroups.
Visual perception and visuospatial abilities
Table 3 provides an overview of the test performances of visual perception and visuospatial abilities. To assess whether NPD groups differ from controls we examined four variables: (1) total raw scores of Cube Analysis, (2) Silhouettes and (3) Copy Task of Visual Reproduction-II; (4) total time of Visual Reproduction-II Copy Task. Results showed significant differences between the group means with large effect size on Cube Analysis, and raw scores of Visual Reproduction-II Copy Task. No significant differences were found in the total time of Visual Reproduction-II Copy Task (p = 0.331).
Table 3. Descriptive statistics for the five groups of test performances of visuospatial abilities and visuospatial perception
Note: The mean, standard deviation, minimum, and maximum values, effect sizes, F-statistic values for ANOVA, and H-statistic values for the Kruskal–Wallis test are reported. U-ADHD = unmedicated ADHD, M-ADHD = medicated ADHD, ASD = autism spectrum disorder, COMBO = both unmedicated ADHD and ASD. Sample sizes differ across groups due to the exclusion of outliers. “KW” indicates p-values calculated using the Kruskal–Wallis test, while ANOVA was used for all other variables. Asterisks (*) denote H-statistic values for the Kruskal–Wallis test. rs = raw scores, sec = seconds, and n = the sample size for each group.
We found after Bonferroni correction significant differences between controls and all four neuropsychiatric groups for Cube Analysis and Copy Task of Visual Reproduction-II. Additionally, we found significant differences only in Silhouettes scores between controls and U-ADHD (p < .001) and controls and ASD (p=.002). No other significant differences emerged between the remaining clinical subgroups.
A two-way two-tailed ANOVA was conducted to explore the impact of total time and diagnosis on total raw score of the Visual Reproduction-II Copy Task. Participants were divided into three groups according to how fast they drew all five figures (Group 1 (n = 58): <109 s, Group 2 (n = 114): between 109 and 192 s, Group 3 (n = 61): >192 s). The interaction effect between total time and diagnosis was not significant (F = 0.79, p = 0.61, η2 = 0.02). There was a statistically significant main effect for total time with medium effect size (F = 14.02, p< 0.001, η2 = 0.114) and a statistically significant main effect for clinical group with large effect size (F = 16.43, p< 0.001, η2 = 0.232).
Cerebellar tests
Table 4 provides an overview of performances on cerebellar tests. To assess whether NPD differ from controls on cerebellar functions, we examinate six variables: (1) Finger Tapping Production Mean; (2) Finger Tapping Local Variation; (3) Finger Tapping Drift; (4) the error on the first trial after taking the prisms on (Prism With); (5) the first trial after taking the prisms off (Prism First Without); and (6) average error across all trials of Prism Adaptation.
Table 4. Descriptive statistics for the five groups of performances on cerebellar tests
Note: The mean, standard deviation, minimum, and maximum values, effect sizes, F-statistic values for ANOVA, and H-statistic values for the Kruskal–Wallis test are reported. U-ADHD = unmedicated ADHD, M-ADHD = medicated ADHD, ASD = autism spectrum disorder, COMBO = both unmedicated ADHD and ASD. Sample sizes vary across groups due to the exclusion of outliers. Data for the Finger Tapping test are missing due to problems with the device. “KW” indicates p-values calculated using the Kruskal–Wallis test, while ANOVA was used for all other variables. Asterisks (*) denote H-statistic values for the Kruskal–Wallis test. ms = milliseconds, cm=centimeters, and n = the sample size for each group.
Results showed significant differences between the group means with large effect size on Prism Error Mean, medium effect size on Prism First With, and with small effect size on Prism First Without and Finger Tapping Production Mean. No significant differences were found for Finger Tapping Local Variation (p = 0.211) and Finger Tapping Drift (p = 0.103).
We found after Bonferroni correction significant differences between control and all four neuropsychiatric groups for Prism First With and Prism Error Mean. Additionally, we found significant differences only in the Finger Tapping Production Mean scores between the controls and the U-ADHD (p = 0.005) and controls and M-ADHD (p = 0.013). For Prism First Without we did not find differences between groups. No other significant differences emerged between the remaining clinical subgroups.
ROC analysis
ROC analysis was performed to investigate the sensitivity and specificity of the tests in predicting neuropsychiatric diagnoses. ROC curves and area under the curve (AUC) values for the variables with high AUC (above 0.80) for all groups are shown in Figure 1. We created a new combined variable which was defined as: Coding + Perception Time + Prism Error Mean. These tests were chosen because they maintain an AUC >0.80 in all four NPD groups and have a large effect size (>0.14). This derived variable yielded an AUC of 0.935 for the ASD group, 0.900 for U-ADHD, 0.874 for M-ADHD and 0.919 for COMBO.
Regression analysis
Linear and multiple regression analyses were used on all participants to control for age, education level, sleep, anxiety, and depression, on neurocognitive outcomes, specifically on Coding, Perception Time, Prism Error Mean and Copy Task of Visual Reproduction-II. Age exhibited small significant associations with Perception Time of the Bender-II Test and Prism Error Mean.
Overall, education consistently exhibited significant associations with neurocognitive outcomes. Results showed that insomnia scores had a mild yet significant impact on performance, depression showed no effect, and anxiety influenced only the processing speed in the Perception Time of the Bender-II Test. The results are presented in Table 5.
Table 5. Results of the regression for the impact of age, education, sleep, anxiety and depression on neurocognitive outcomes
Note: This table presents the results of linear and multiple regression analyses examining the influence of age, education, sleep (ISI scores), anxiety (HADA scores), and depression (HADD scores) on four dependent variables: Copy Task of Visual Reproduction-II, Coding, Perception Time, and Prism Error Mean. The table includes R2 values, F values with degrees of freedom (df), standardized regression coefficients (β), standard errors (SE), 95% confidence intervals (CI) for β, and p-values.
Linear regression analyses also were conducted to investigate the relationships between sensorimotor function tests, serving as independent variables, and the Copy Task of Visual Reproduction-II, which measures visuospatial abilities, as well as Cube Analysis, which assesses visuospatial perception, as dependent variables. The aim was to explore the potential negative impact of sensorimotor deficits on visuospatial perception and visuospatial abilities.
Coding is positively associated with performance on the Copy Task of Visual Reproduction-II, with a significant regression coefficient, explaining 15.4% of the variance. Coding also has a positive association with Cube Analysis performance, explaining 13.9% of the variance.
The exploratory variable Perception Time negatively affects Copy Task performance, indicating that longer perception times are associated with poorer outcomes. This accounts for 11.1% of the variance. Perception Time has a significant negative impact on Cube Analysis outcomes, with slower perception times linked to poorer performance. This predictor explains 19.0% of the variance. This reinforcing the notion that sensorimotor deficits, particularly slower Perception Time, can reduce outcomes in visual perception and visuospatial abilities. The results are presented in table 6.
Table 6. Results of regression for the impact of sensorimotor deficits on visuospatial perception and visuospatial abilities
Note: This table summarizes the results of regression analyses examining the impact of sensorimotor deficits on visuospatial perception and visuospatial abilities. The table includes R2 values, F values with degrees of freedom (df), standardized regression coefficients (β), standard errors (SE), 95% confidence intervals (CI) for β, and p-values.
Discussion
The main aim of this study was to measure sensorimotor functions that require cerebellar processing, visuospatial perception, and visuospatial abilities between control and NPD groups, to examine subgroup differences in four NPD clinical groups using objective tests for these functions. Our results indicate that all NPD groups have lower performance on almost all variables measuring those functions compared to controls.
For sensorimotor functions, there were significant differences with a large effect size in RTIFMMT of Reaction Time between controls and U-ADHD, ASD, and COMBO, and as well as between ASD and M-ADHD. For MOTML scores of MOT, significant differences with a medium effect size were found between controls and ASD, and between M-ADHD and ASD. RTIFMRT scores of Reaction Time showed significant differences with a small effect size only between controls and U-ADHD. The results indicate differences in sensorimotor function between NPD groups and controls, with the ASD group showing the largest deviation. These findings suggest that individuals in NPD groups face sensorimotor challenges, which may lead to difficulties in tasks requiring fine motor control and problems coping with overwhelming environments.
For visuospatial abilities, there were no differences between the clinical subgroups but there were differences between controls and clinical subgroups. However, for visuospatial perception, there were significant differences in performance on the Silhouettes of the VOSP. A medium effect size was found, with significant differences observed between the control group and U-ADHD, as well as between the control group and ASD. The results show that clinical subgroups exhibit altered visuospatial processing compared to controls. When comparing the different clinical subgroups, we found variations in visuospatial perception but not in visuospatial abilities. These difficulties may manifest in various aspects of patients’ daily lives. For instance, individuals might struggle with tasks that require accurate interpretation of visual details, such as reading, recognizing faces, or identifying objects in cluttered environments. Additionally, the altered visuospatial processing could affect their ability to navigate their surroundings, judge distances, or perform tasks like driving or organizing physical spaces. These challenges may lead to increased frustration or dependency on others for activities that typically rely on intact visual and spatial skills.
For cerebellar tests, there were differences in the Production Mean of Finger Tapping with a small effect size between the control group and both U-ADHD and M-ADHD. This indicates U-ADHD and M-ADHD groups exhibit subtle impairments in motor timing or coordination. In everyday life, the subtle impairments in motor timing and coordination may translate to challenges in activities that require precise and consistent motor control. These challenges can also affect rhythm-based activities, potentially causing frustration or impacting academic, occupational, and social functioning. These differences could help guide more targeted diagnostic assessments focusing on sensorimotor functions, visuospatial perception, visuospatial abilities and cerebellar tests within these populations.
In addition, our results showed that higher sensorimotor function is strongly associated with better performances in visual perception and visuospatial tasks, underscoring its role in these cognitive functions. In addition, slower perception time is associated with the outcomes, highlighting the detrimental impact of sensorimotor deficits. These findings emphasize the essential contribution of sensorimotor functions in facilitating visuospatial and visual perceptual abilities.
This study also aimed to investigate the impact of central nervous system (CNS) stimulants on cognitive functions in adults with ADHD. We found that individuals with U-ADHD but not M-ADHD exhibited slower reaction times, lower accuracy, motor coordination deficits compared to controls. In contrast, M-ADHD demonstrated higher accuracy and faster completion time. These results suggest that medication has a positive effect on cognitive functions, aligning with previous findings (Low et al., Reference Low, Vangkilde, le Sommer, Fagerlund, Glenthøj, Jepsen, Bundesen, Petersen and Habekost2019). Overall, medication appears to mitigate some cognitive and motor deficits in ADHD, particularly in reaction time and visuospatial perception tasks, with U-ADHD individuals showing more pronounced impairments. This underscores the importance of medication in addressing sensorimotor and cognitive challenges in ADHD.
The results on the WAIS-IV Coding subtest with a large effect size align with previous findings, indicating that patients with ADHD score lower than controls on tasks requiring graphomotor demands (Becker et al., Reference Becker, Marsh, Holdaway and Tamm2020). This test also measures incidental learning ability, which is reduced in these clinical populations. Our findings are in line with previous studies (Braconnier & Siper, Reference Braconnier and Siper2021; Marinopoulou et al., Reference Marinopoulou, Lugnegård, Hallerbäck, Gillberg and Billstedt2016; Mohamed et al., Reference Mohamed, Butzbach, Fuermaier, Weisbrod, Aschenbrenner, Tucha, Tucha and Garcia2021).
On the exploratory time variable for the Perception Time of Bender-II, we observed significant differences between all four NPD groups and controls, with high AUC values and with a large effect size. This test exhibits strong psychometric properties. In our sample, there was a strong concurrent validity between Perception Time and Symbol Search, as well as strong split-half reliability. Conversely, there were no differences in accuracy. Taken together, all groups of patients display sensorimotor deficits. This is consistent with cerebellar involvement in these neuropsychiatric disorders (Stoodley, Reference Stoodley2014, Reference Stoodley2016).
In our sample, the timed visual copy task showed a large effect size. Previous studies using Visual Reproduction-II have not looked at Copy Task accuracy (Hida et al., Reference Hida, Hayashi, Okajima, Takashio and Iwanami2020; Müller et al., Reference Müller, Gimbel, Keller-Pliessnig, Sartory, Gastpar and Davids2007), which, according to our results, can help distinguish between controls and NPD groups. Importantly, the Visual Reproduction-II Copy Task was developed to assess visuomotor integration and ability of a subject as a confounder in the assessment of visual memory (Tulsky et al., Reference Tulsky, Saklofske, Chelune, Heaton, Ivnik, Bornstein, Prifitera and Ledbetter2003).
Two previous studies using Rey–Osterrieth Complex Figure Test did not find any significant difference in copy time or accuracy between late-onset ADHD, ADHD diagnosed in childhood, subthreshold ADHD, and the control group (Faraone et al., Reference Faraone, Biederman, Doyle, Murray, Petty, Adamson and Seidman2006; Rapport et al., Reference Rapport, Van Voorhis, Tzelepis and Friedman2001). Another study using the Rey–Osterrieth Complex Figure Test copying condition did find a significant difference in accuracy where the ADHD group performed worse (Schreiber et al., Reference Schreiber, Javorsky, Robinson and Stern1999). The time to copy the Rey–Osterrieth Complex Figure Test was significantly slower in adults with ASD but no differences were found in accuracy (Davids et al., Reference Davids, Groen, Berg, Tucha and van Balkom2020). In contrast to previous studies (Aycicegi-Dinn et al., Reference Aycicegi-Dinn, Dervent-Ozbek, Yazgan, Bicer and Dinn2011; Callahan et al., Reference Callahan, Ramakrishnan, Shammi, Bierstone, Taylor, Ozzoude, Goubran, Stuss and Black2022; Minshew et al., Reference Minshew, Goldstein and Siegel1997; Wilson et al., Reference Wilson, Happé, Wheelwright, Ecker, Lombardo, Johnston, Daly, Murphy, Spain, Lai, Chakrabarti, Sauter and Murphy2014), we observed clear deficits in visuospatial abilities. Nevertheless, it is important to take into account that the deficits in visual perception and visuospatial abilities could be partly due to executive dysfunctions (Schreiber et al., Reference Schreiber, Javorsky, Robinson and Stern1999).
We observed a clear difference with a large effect size between controls and NPD groups for Prism Adaptation which requires cerebellar activation and assesses both sensorimotor functions and the adaptation of motor activity in response to changes in visual input (Küper et al., Reference Küper, Wünnemann, Thürling, Stefanescu, Maderwald, Elles, Göricke, Ladd and Timmann2014). This is in line with observations in our previous study (Gustafsson et al., Reference Gustafsson, Kjell, Cundari, Larsson, Edbladh, Madison, Kazakova and Rasmussen2023). For Finger Tapping, which has been linked to cerebellar function (Rivkin et al., Reference Rivkin, Vajapeyam, Hutton, Weiler, Hall, Wolraich, Yoo, Mulkern, Forbes, Wolff and Waber2003), we observed a significant small effect size in production mean but unlike in our previous study (Gustafsson et al., Reference Gustafsson, Kjell, Cundari, Larsson, Edbladh, Madison, Kazakova and Rasmussen2023), we did not see a difference in tapping variability (Drift and Local). Possible reasons for this could be the lower severity index, or differences in the medication scheme.
Although the majority of the patients tested had a mild severity index (60.6%–83.9%), we observed significant results. Typically, individuals with severe or moderate ADHD or ASD receive their diagnosis in childhood. It is likely that the deficits we observed are more pronounced in patients with a moderate or severe form.
In interpreting our findings, it is important to consider the potential impact of medication and comorbidities on the differences observed between ADHD and ASD. Given the frequent presence of anxiety, depression, and sleep disorders in these conditions, as well as the widespread use of medication, it is challenging to isolate the effects of the neuropsychiatric disorders themselves. This aligns with previous findings highlighting the high heterogeneity of psychiatric comorbidities among NPD patients (Katzman et al., Reference Katzman, Bilkey, Chokka, Fallu and Klassen2017). However, in clinical practice, patients with ADHD or ASD are rarely unmedicated, making it difficult to draw definitive conclusions about the role of these factors in cognitive outcomes. Our regression analysis revealed that only insomnia had a significant impact on cognitive performance. Depression did not influence performance on the cognitive tests, while anxiety only influenced processing speed in the Perception Time of the Bender-II Test. The diversity of medications used in our sample further complicates the ability to assess their specific effects on cognitive performance. More research with larger and more homogenous samples is needed to better understand how these factors interact.
Cerebellar function in NPD patients
Our test battery included Finger Tapping and Prism Adaptation – two tests linked to cerebellar function. The results indicate NPD patients with predominantly mild degree of severity have deficits on these cerebellar tasks even though. This reinforces the view that cerebellar deficits are a component of these disorders. Although our study was guided by theoretical models and research on the cerebellum’s involvement in neuropsychiatric disorders and its connections with sensorimotor integration, visuospatial perception, and visuospatial abilities, we cannot draw specific conclusions about these connections due to the absence of neuroimaging data. However, based on previous research, we hypothesize that the cerebellum plays a significant role in these processes (Stoodley, Reference Stoodley2014, Reference Stoodley2016).
The cerebellum is crucial for coordinating movements and is implicated in visuospatial processing. It has been linked to dysmetria – a condition where individuals struggle to control the range of their movements (Bastian, Reference Bastian2002). This dysfunction can affect tasks requiring precise motor control, such as visual copy tasks (Slapik et al., Reference Slapik, Kronemer, Morgan, Bloes, Lieberman, Mandel, Rosenthal and Marvel2019). In our study, we observed that patients, compared to controls, exhibited significant difficulties with figure organization, proportions, and placement in the Copy Task of Visual Reproduction-II.
Previous work has addressed the role of the posterior parietal cortex in drawing, but fMRI studies have shown that it also activates the cerebellum (Bai et al., Reference Bai, Liu and Guan2021). The cerebellum acquires and maintains an internal forward model that predicts current and future states of the body by utilizing using efference copies of motor commands.
The drawing patterns we observed in the patients are consistent with mild dysmetria. For example, some of the drawn figures were much larger or smaller than the reference, which could occur if the patient consistently overshoots or undershoots in their drawing. Our results thus align with the idea that impairment in the predictive computation for voluntary movements accounts for many characteristics associated with dysmetria (Cabaraux et al., Reference Cabaraux, Gandini, Kakei, Manto, Mitoma and Tanaka2020).
Cerebellar contributions to cognitive processing have been less studied in ADHD and ASD. Other studies have shown that the cerebellum and the hippocampus interact during spatial navigation (Rochefort et al., Reference Rochefort, Lefort and Rondi-Reig2013). The cerebellum is thought to be involved in the precise encoding and computation of self-motion information from various sources, which is necessary for constructing the body’s representation in space (Rochefort et al., Reference Rochefort, Lefort and Rondi-Reig2013). Still, more research is necessary to investigate the cerebellum’s role in visuospatial processing, even more so when relating to ADHD and ASD. Patients with cerebellar ataxia have been shown to exhibit deficits in Copy Task of Rey–Osterrieth Complex Figure Test (Slapik et al., Reference Slapik, Kronemer, Morgan, Bloes, Lieberman, Mandel, Rosenthal and Marvel2019), mirroring findings of significantly lower scores for ADHD subjects on Rey–Osterrieth Complex Figure Test Copying organization (Faraone et al., Reference Faraone, Biederman, Doyle, Murray, Petty, Adamson and Seidman2006). As mentioned above, studies examining the connections between the cerebellum, visuospatial abilities in ADHD and ASD are sparse.
There are, however, indications that visuospatial abilities, especially visuospatial construction, are affected in ADHD (Schreiber et al., Reference Schreiber, Javorsky, Robinson and Stern1999). There is also evidence that the cerebellum has a role in the functional differences in ADHD and ASD as well as visuospatial functioning in general (Cundari et al., Reference Cundari, Vestberg, Gustafsson, Gorcenco and Rasmussen2023; King et al., Reference King, Hernandez-Castillo, Poldrack, Ivry and Diedrichsen2019; Stoodley, Reference Stoodley2014, Reference Stoodley2016). We speculate that challenges in visuospatial ability may partly stem from cerebellar-related sensorimotor deficits, which are shared in patients with ADHD and ASD, even when the severity is mild.
Limitations
Patients with NPD are a highly heterogeneous group. Sleep disorders, anxiety, and depression co-occur creating a complex interplay of symptoms. Pharmacological treatment and active metabolites can also act as confounding variables. Here we did not ask the patients to stop their pharmacological treatment to participate in the study. Presumably, the performance of patients on medication would have been poorer if they did not take medication. The patients were diagnosed by different psychiatrists and psychologists. In this study, the severity index was estimated based on the patient’s clinical symptoms and functionality according to the DSM-V guidelines. Yet the lack of standardized assessment tools to estimate the severity index makes it difficult to accurately evaluate this variable.
Conclusion
This study highlights deficits in sensorimotor functions, visuospatial perception, and visuospatial abilities in NPD groups compared to controls. Significant correlations were found between sensorimotor functions, visuospatial perception, and visuospatial abilities. Post hoc analysis revealed differences between NPD groups, particularly between M-ADHD and U-ADHD, underscoring the impact of CNS stimulants. Our results suggest that the Copy Task of Visual Reproduction-II of WMS-III, Prism Adaptation, Coding of WAIS-IV, and Perception Time of Bender-II can differentiate controls from patients with ADHD, ASD, or both. These tests have the potential to serve as screening tools or as support for the diagnosis together with other DSM-V criteria.
Future research
Future research should investigate neural activity during sensorimotor, visual-perceptive, and visuospatial tasks in children and adults with NPD. The findings need to be replicated in new samples. Neuroimaging studies are necessary to examine the extent of cerebellar activity related to visuospatial perception and visuospatial abilities. To evaluate the clinical efficacy of these tests, it is crucial to assess a diverse range of psychiatric diagnoses. This is essential for conducting comprehensive differential diagnoses.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1355617725000189.
Data availability
The data that support the findings of this study are available at reasonable request from the corresponding author MC. The data are not publicly available due to them containing information that could compromise research participant privacy.
Acknowledgements
We thank the patients and the controls that participated in this study. MC thanks primarily Gudrun Olow Schnaars (previous manager of the unit of adult neuropsychiatry), Rosie Setälä (current manager) and Ewa Hellberg (department coordinator) which understood, facilitated the research procedures and inclusion of patients at the unit of adult neuropsychiatry in Helsingborg, Sweden.
Author contribution
MC: designed the test battery with SV, tested all the patients and most of the controls, did statistical analysis, wrote the first draft, conception, disposition and organization of the work. SV: reviewed and critiqued all the work. AH: tested some of the controls, corrected most of the test results and did statistical analysis. JK: wrote part of the first draft inherent to ADHD. PG: reviewed and critiqued the clinical aspect inherent to ADHD and ASD. AR: supervised the main work, conception, financing, and organization of the work. All authors contributed to the article and approved the submitted version.
Funding statement
This study was supported by grants to Anders Rasmussen from the Swedish Research Council (2020-01468), the Crafoord Foundation (20180704, 20200729, 20220796, 20230655), The Brain Foundation (FO2024-0361), Fysiografiska sällskapet (2022-04-25), Gyllenstiernska Krapperupsstiftelsen (KR2021-0040 & KR2024-0102) the Segerfalk foundation (2019-2246), Åke- Wibergs foundation (M18-0070 & M19-0375, M20-0008), Petrus och Augusta Hedlunds Foundation (M-2024-2498), Fredrik & Ingrid Thurings foundation (2018-00366 & 2019-00516), Pia Ståhls Foundation (20,012), Magnus Bergvalls Foundation (2020-03788 & 2023-861), and Sten K Johnsons foundation (20230153 & 20240482). The funders had no role in study design, data collection, analysis, decision to publish, or manuscript preparation.
Competing interests
No competing interests.
Ethical approval and consent to participate
This study was a part of a research project approved by the Swedish Ethical Review Authority (Dnr 2022-01799-01). Participants were informed that their participation was voluntary and could be terminated at any point. A written consent form was given to the participant to confirm the conditions regarding the use of collected data. All participants were assigned a non-identifying code. The participants were informed that they could refuse to answer any question in case the topic was sensitive. If a participant exhibited symptoms of a more serious mental disorder, suicidal thoughts, or if results on any neuropsychological test indicated any possible need for healthcare the participant would be referred to an appropriate resource.
Consent to publication
All participants were informed that the results would be published but that they would remain anonymous. The manuscript contains no identifiable information.
Open access
