Introduction
Method
Search Strategy
Inclusion Criteria
Procedure
Data Extraction and Synthesis
First author | Year | Method | Studies /type | Sample | Mean age/years | Quality | Results in comparison to neurotypical peers |
---|---|---|---|---|---|---|---|
Alderson | (2013) | SR MA | 38 N | ADHD n = 1338 Controls n = 1393 | Adults | Mod | Adults with ADHD displayed atypical phonological and visuospatial working memory (moderate effect size), indicating working memory problems persist beyond childhood. Several task-moderating variables explained significant effect size variability (heterogeneity) among studies. Risk of publication bias was non-significant |
Alderson | (2007) | SR MA | 25 N | ADHD n = 808 Controls n = 695 | Children | Mod | Children with ADHD displayed significantly slower mean reaction time, greater reaction time variability, and slower stop-signal reaction time (moderate effect size). Findings pertain to executive-motor inhibition. Several subject and task variables served as significant moderators for children’s mean reaction time (heterogeneity). Risk of publication bias not assessed |
Aoki | 2018 | SR MA | 27 P | ADHD n = 871 Controls n = 846 | Children Adolescents Adults | Mod | TBSS: children and adults with ADHD display significantly lower fractional anisotropy values in four clusters (two in the corpus callosum, one in R inferior fronto-occipital fasciculus, and one in L inferior longitudinal fasciculus). WBVBA: elevated fractional anisotropy in three clusters (L mid-cingulate, extending to the corpus callosum, anterior corpus callosum, and L inferior fronto-occipital fasciculus). Results from different methodology did not converge and many studies did not examine potential group differences in head motion. In studies that did, most reported no significant results. Risk of publication bias not assessed |
Aoki | 2013 | MA | 16 P | ADHD n = 270 Controls n = 235 | Children Adolescents Adults | High | Childhood ADHD is associated with significantly higher-than-normal N-acetylaspartate, a marker of neuronal activity, in the medial prefrontal cortex. No significant difference was found in adults with ADHD and meta-regression revealed a significant correlation between advanced age and normal levels of N-acetylaspartate in the ADHD cohort. Results did not reveal publication bias and heterogeneity was negligible |
Arns | 2013 | MA | 9 P | ADHD n = 1253 Controls n = 517 | Children Adolescents | Low | Children with ADHD displayed elevated theta/beta ratio (moderate effect size). Results were marred by significant heterogeneity and are therefore considered misleading. Risk of publication bias not assessed |
Arora | 2020 | MA | 9 N | ADHD n = 491 Controls n = 402 | Children | Low | Children with ADHD displayed atypical alerting and executive attention networks. Non-zero heterogeneity. Risk of publication bias not accessed |
Balogh | 2016 | MA | 15 P | ADHD n = 1053 Controls n = 614 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed significantly diminished post error slowing after committing an error (medium effect size), regardless of gender. They also tended to sustain or even increase the speed of their reactions post error. Risk of publication was non-significant |
Bellato | 2020 | SR | 55 P | ADHD n = > 1400 Controls n = > 1400 | Children Adolescents Adults | Low | Children and adults with ADHD tended to display atypical autonomic nervous system arousal, more often in the direction of hypo-arousal than hyper-arousal, particularly at rest and during tasks requiring response regulation and sustained attention. Half the reported findings were null. Publication bias not assessed. It is too early to say whether atypical autonomic nervous system dysfunction is related to ADHD as a continuous trait measure |
Boedhoe | 2020 | C/S ENIGMA | - | ADHD n = 2271 ASD n = 1777 OCD n = 2323 Controls n = 5827 | Children Adolescents Adults | High | Children and adolescents with ADHD displayed reduced intercranial volume, reduced frontal surface area and striatal volume in comparison (small to moderate effect size). None of the subcortical volumes differed significantly among adult patient groups |
Boon | 2020 | SR | 174 P N | n = > 8000 | Children Adolescents | Low | Children and adolescents with ADHD displayed significant differences in neural anatomy, connectivity and innovation, and cognitive functioning. Publication bias not assessed |
Boonstra | 2005 | MA | 13 N | ADHD n = 662 Controls n = 417 | Adults | Low | Adults with ADHD displayed atypical verbal fluency, inhibition, and set-shifting, as well as atypical reaction time (moderate effect size). Heterogeneity and risk of publication bias not accessed |
Chamorro | 2022 | MA | 31 P | n = 1567 | Adolescents Adults | High | Individuals with ADHD commit more ocular inhibition errors in the form of anticipatory rapid eye movement responses during memory guided saccade task waiting periods (moderate to large). Risk of publication bias corrected during meta-analysis. Results suggests that the capacity to withhold a response on presentation of a relevant stimulus is impaired in adolescents and adults with ADHD. Due to significant heterogeneity results should be interpreted with caution |
Chen | 2023 | MA | 16 N | ADHD n = 2928 Controls n = 2775 | Children Adolescents Adults | High | Children and adults with ADHD consistently displayed significantly smaller amygdala surface area (particularly in the L hemisphere), without a significant difference in volume. The effects of the scanner types, segmentation methods, and the investigated continuous variables (age, IQ, and the male percentage) on ADHD-associated anatomical alterations of the amygdala size were nonsignificant. Risk of publication bias was non-significant |
Chen | 2016 | SR MA | 10 P | ADHD n = 470 Controls n = 479 | Children Adolescents Adults | High | Individuals with ADHD displayed significant fractional anisotropy reductions in the commissural fibres that connect bilateral hemispheres; the association fibres that link occipital, frontal and temporal regions; and projection fibres that pass through the corona radiata and cerebellum. The most evident and consistent white matter differences were located in the splenium of the corpus callosum extending to the R cingulum, the R sagittal stratum and L tapetum. The projection fibres that pass through the corona radiata and cerebellum may also be affected, although sensitivity analysis failed to retain significance. Significant between-group heterogeneity was detected in the splenium of the corpus callosum and the R sagittal stratum. Results were free from publication bias and indicate that the interruptions in the white matter tracts are related to interhemispheric communication, posterior brain circuitries and the limbic system |
Cheng | 2016 | SR MA | 6 P | ADHD n = 73 Controls n = 75 | Children Adolescents | High | Children and adolescents with ADHD demonstrated significantly smaller mismatched negativity amplitudes, indicating less vigilance to auditory change, reduced involuntary attention switching, and poor auditory sensory memory—a component of working memory (small affect size). No statistical evidence was found for publication bias or heterogeneity |
Connaughton | 2022 | SR | 46 P | ADHD n = 1589 Controls n = 1410 | Children Adolescents | Mod | Children and adolescents with ADHD tended to display widespread atypical white matter microstructure in both discrete white matter tracts and neural networks (mostly decreased connectivity but also increased in some areas). The most prominent differences were found in the fronto-striatal tracts, corpus callosum, superior longitudinal fasciculus, cingulum bundle, thalamic radiations, internal capsule and corona radiata. Many of the affected neural networks are associated with key neuropsychological functions that are atypical in ADHD. Heterogeneity in the literature may stem from a variety of methodological limitations. Risk of publication bias not assessed |
Cortese | 2021 | SR MA | 30 P | ADHD n = 1094 Controls n = 884 | Children Adolescents Adults | Mod | Although studies report ADHD-related hyperconnectivity and hyperconnectivity, no spatial convergence was achieved. This may be due to heterogeneity in study participants, methodology, experimental procedures, and analytic flexibility as well as in ADHD pathophysiology. Post hoc meta-analysis showed ADHD is associated with consistently dysregulated L superior temporal gyrus, which is involved in auditory processing and social cognition. Risk of publication bias not assessed |
Cortese | 2016 | MA | 24 P | ADHD n = 422 Controls n = 433 | Adults | Low | Adults with ADHD displayed hypoactivation in the L putamen, L inferior frontal gyrus (pars opercularis), L temporal pole, and R caudate. The L inferior gyrus and R caudate mapped to executive function tasks. When focusing on studies including participants free of comorbidity, only one region (L putamen) was found to be consistently hypoactivated. Risk of publication bias not assessed |
Cortese | 2012 | SR MA | 55 P | ADHD n = ≈ 741 Controls n = ≈ 801 | Children Adolescents Adults | High | Children with ADHD displayed significant (1) hypoactivation in the frontal, R parietal and R temporal regions, putamen bilaterally and in the systems that govern executive function (frontoparietal network) and attention (ventral attentional network), and (2) hyperactivation in the R angular gyrus, middle occipital gyrus, posterior cingulate cortex, and midcingulate cortex and in the default, ventral attention, and somatomotor networks. Stimulant-naïve children displayed hypoactivation in the R superior temporal gyrus, bilateral putamen and R thalamus. Adults with ADHD displayed significant (1) hypoactivation in the R central sulcus, precentral gyrus, and middle frontal gyrus, and frontoparietal system; and (2) hyperactivation in a region with a peak in the R angular and middle occipital gyri, and in the visual, dorsal attention, and default networks. ADHD-related dysfunction largely reflected task features and was detected even in the absence of comorbid mental disorders or a history of stimulant treatment. Risk of publication bias not assessed |
Cortese | 2006 | SR MA | 13 N | ADHD n = 280 Controls n = 228 | Children Adolescents | Mod | Children with ADHD display significantly reduced mean sleep latency (large effect size), fall asleep more often during multiple sleep latency testing (large effect size), and may display excessive daytime sleepiness which impacts alertness. Children with ADHD display significantly more movement during sleep, a significantly higher apnoea-hypopnea index and sleep-disordered breathing. Risk of publication bias not assessed |
Cui | 2023 | MA | 22 P | ADHD n = 1158 Controls n = 1204 | Children Adolescents Adults | High | Individuals with ADHD displayed reliable and robust age-related significantly reduced fractional anisotropy in the splenium of the corpus collosum. The adult ADHD subgroup also displayed reduced fractional anisotropy in the body of the corpus collosum, indicating white matter tract integrity reduces with increasing age. No clear evidence of between-study heterogeneity or publication bias identified |
Dan | 2020 | IR | 9 P N | ADHD n = 313 Controls n = 327 | Children Adolescents | Very Low | Adolescents with ADHD display no differences in reaction time, reaction time variance and cognition accuracy when recognising emotional facial expressions, however, brain activity and temporal evolution differences are present when neural responses are recorded using f-MRI or event related potentials. Risk of publication bias not assessed |
Dekkers | 2016 | SR MRA | 37 N | ADHD n = 1175 Controls n = 1222 | Children Adolescents Adults | High | ADHD is related to significantly increased risky decision making on gambling tasks in laboratory settings (small to medium effect size). Larger effect sizes were associated with co-morbid Disruptive Behaviour Disorder. Age did not moderate outcomes. No evidence of publication bias |
Dickstein | 2006 | MA | 16 P | ADHD n = 184 Controls n = 187 | Children Adolescents Adults | Low | Across studies, individuals with ADHD displayed significant patterns of frontal hypoactivity within the anterior cingulate, dorsolateral prefrontal, inferior prefrontal, and orbitofrontal cortices, as well as related regions, such as portions of the basal ganglia and parietal cortices. When focusing on studies of response inhibition alone, a more limited set of group differences were observed, including inferior prefrontal cortex, medial wall regions (including anterior cingulate cortex) and the precentral gyrus. In contrast, analyses focusing on studies of constructs other than response inhibition revealed a more extensive pattern of hypofunction in patients with ADHD than those of response inhibition. Risk of publication bias not assessed |
Doidge | 2021 | MA | 28 N | ADHD n = 1055 Controls n = 3485 | Children Adolescents Adults | Mod | During temporal discounting and delay of gratification tasks, females with ADHD chose smaller immediate rewards over larger delayed rewards more often than males with ADHD. In contrast, males with ADHD were more likely to choose the larger delayed reward. Task type, age, and reward type did not significantly predict sex differences. Heterogeneity across studies was low and non-significant and no evidence of publication bias was found |
Dutra | 2016 | SR MA | 9 P | ADHD n = 217 Controls n = 244 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed short intracortical inhibition (a marker of motor inhibition). No differences in resting motor threshold and silent period were found. Significant statistical heterogeneity was detected between trials. Clinical and methodological heterogeneity may also have been present. Risk of publication bias not assessed |
Ellison-Wright | 2008 | MA | 7 P | ADHD n = 114 Controls n = 143 | Children Adolescents Adults | Low | Individuals with ADHD displayed significant regional grey matter reduction the R putamen/globus pallidus region. Clinical heterogeneity and publication bias may be present; therefore, the possibility of grey matter decreases in the bilateral putamen/globus pallidum, caudate and frontal lobes cannot be ruled out. Risk of publication bias not assessed |
Frodl | 2012 | SR MA | 11 P | ADHD n = 320 Controls n = 288 | Children Adolescents Adults | Mod | Children with ADHD displayed significantly reduced grey matter volumes in the R globus pallidus, R putamen and decreased bilateral caudate volumes. Adults with ADHD displayed significantly reduced anterior cingulate cortex volumes. Meta-regression indicated that studies in untreated children may also show changes in the anterior cingulate cortex and the amygdala region. Heterogeneity was not seen on the putamen and anterior cingulate cortex but was present in the globus pallidus and in the frontal, temporal, and parietal lobes that did not show significant differences. Risk of publication bias not assessed |
Fusar-Poli | 2012 | MA | 9 P | ADHD n = 169 Controls n = 173 | Adults | Mod | Although striatal dopamine transporter density was 14% higher on average in the ADHD group, heterogeneity across studies was large and statistically significant. Pooled meta-analysis indicated consistent statistical evidence for greater dopamine transporter density in the ADHD group in the whole striatum (small effect size). Meta-regression analyses showed that the percentage of subjects without exposure to psychostimulants was negatively correlated with dopamine transporter density; density was higher in patients with previous medication exposure and lower in medication-naive patients. There was no moderating effect for age, comorbidity, gender, year of publication, or imaging technique. No evidence of publication bias found. Sensitivity analysis confirmed robustness of the results |
Gao | 2019 | SR MA | 21 P | ADHD n = 700 Control n = 580 | Children Adolescents Adults | Mod | ADHD is characterised by hyperconnectivity between the frontoparietal network and regions of the default node network and affective network, and hypoconnectivity between the frontoparietal network and regions of the ventral attention network and somatosensory network. No significant between-group heterogeneity was found in the results for the default mode network and affective network. For the frontoparietal network results, significant between-group heterogeneity was detected in the superior frontal gyrus and the putamen. No evidence of publication bias found |
Geburek | 2013 | MA | 7 P | ADHD n = 166 Controls n = 161 | Adolescents Adults | Low | Error negativity was significantly attenuated in adolescents and adults with ADHD when participating in event-related potential and behaviour performance tasks (medium effect size). Task type, age and proportion of males did not affect results and low heterogeneity was noted. Risk of publication bias not assessed. Error positivity was attenuated only during GoNogo tasks, however significant heterogeneity was detected. Adolescents and adults with ADHD responded slower during flanker tasks and made more errors during GoNogo and flanker tasks |
Geiss | 2023 | SR | 15 P | ADHD n = 424 Controls n = 422 | Adults | Mod | Individuals with ADHD displayed atypical cardiovascular autonomic modulation, predominantly in the form of reduced sympathetic modulation, during tasks that require attention and/or response regulation or induce emotional stress. Clinical and methodological heterogeneity may influence results. Risk of publication bias not assessed |
Groen | 2013 | SR | 25 N | ADHD n = 257 Controls n = 253 | Children Adolescents Adults | Low | Children and adults with ADHD often displayed increased risky performance during gambling tasks (the effect size ranged between small and large depending on the task). The risk is stronger for children/adolescents with ADHD than for adults with ADHD. Results were highly heterogenous. Comorbid ODD/CD appears to increase the risk. Risk of publication bias not assessed |
Hart | 2013 | SR MA | 34 P | ADHD n = 458 Controls n = 498 | Children Adolescents Adults | Low | Children and adults with ADHD displayed significantly reduced activation for (1) inhibition in the R inferior frontal cortex, supplementary motor area, and anteri- or cingulate cortex, as well as striato-thalamic areas, and (2) attention (interference control) in the R dorso- lateral prefrontal cortex, posterior basal ganglia, and thalamic and parietal regions. Inhibition meta-analysis showed the supplementary motor area and basal ganglia were under activated solely in children with ADHD, while the inferior frontal cortex and thalamus were under activated solely in adults with ADHD. Risk of publication bias not assessed |
Hart | 2012 | MA | 11 P | ADHD n = 150 Controls n = 145 | Adolescents Adults | Mod | Individuals with ADHD displayed significantly reduced activation in typical areas related to timing (i.e., L inferior prefrontal cortex/insula, cerebellum, and L inferior parietal lobe), and in areas that presumably reflect problems with deactivation of the default mode network. Comorbidity did not influence findings. Risk of publication bias not assessed |
Hervey | 2004 | SR MA | 33 N | Unable to determine | Adults | Low | Adults with ADHD displayed atypical neuropsychological functioning, most notably in the domains of attention, distractibility, response inhibition, and memory, consisting of working memory, short-term memory and verbal long-term memory (moderate effect size). A small effect size was also found for processing speed and motor speed. Clinical heterogeneity present. Risk of publication bias not assessed |
Hokken | 2023 | ScR | 35 N | ADHD n = 788 Controls n = 865 | Children | Low | Overall, the literature was found to be inconclusive on accuracy performance during visual search tasks in children with ADHD. However, children with ADHD seem to be more at risk for visual search performance accuracy impairments compared to controls. Risk of publication bias not assessed |
Hoogman | 2019 | C/S ENGIMA | - | ADHD n = 2246 Controls n – 1934 | Children Adolescents Adults | High | Only children with ADHD displayed slightly reduced but statistically significant total cerebral cortex surface area, most notably in the superior frontal gyrus, lateral orbitofrontal cortex, posterior cingulate cortex, rostral anterior cingulate cortex and middle temporal gyrus. The effects were most prominent in the youngest children (4–9 years). Children with ADHD also displayed slightly reduced but statistically significant cortical thickness in 2 regions (fusiform gurus and temporal pole). These differences were most prominent in the group aged 10 and 11 years |
Hoogman | 2017 | MA of C/S data | - | ADHD n = 1713 Controls n = 1529 | Children Adolescents Adults | High | Children with ADHD displayed small but significant reductions in intracranial volume as well as significantly smaller nucleus accumbens, amygdala (largest effect size), caudate nucleus, hippocampus, and putamen volumes. These differences were not present in adults with ADHD. Explorative lifespan modelling suggested a delay of maturation and a delay of degeneration. Psychostimulant medication usage or the presence of comorbid psychiatric disorders did not influence results, nor did symptom scores correlate with brain volume |
Hou | 2023 | SR | 16 P | ADHD n = 321 Controls n = 334 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed smaller, weaker or inactive activation in the frontal cortex during working memory and inhibition control. No publication bias present, however, the following was detected: information bias (two included studies) and confounding bias (two included studies) |
Huang-Pollock | 2012 | SR MA | 47 N | ADHD n = 1181 Controls n = 1518 | Children Adolescents | Mod | School aged children with ADHD displayed difficulty with (1) overall performance on tasks requiring sustained attention (large effect size) and (2) maintaining attention over time (small to medium effect size). They committed more omission and commission errors and displayed slower/more variable reaction times. Measurement unreliability accounted for a significant proportion of reported heterogeneity. Publication bias corrected during meta-analysis |
Huizenga | 2009 | MA | 41 N | ADHD n = 1540 Controls n = 1433 | Children Adolescents | Mod | Inhibitory dysfunction in children and adolescents with ADHD, assessed by the stop-signal reaction time, is moderated by task complexity (stop signal reaction time difference significantly increased with increased with task complexity), and is most pronounced for spatially noncompatible responses. Participant age, medication status, percentage of female participants and number of learning trials did not affect findings. Risk of publication bias not assessed |
Hulsbosch | 2021 | SR | 19 N | ADHD n = 539 ADHD/ODD/CD n = 17 Controls n = 546 | Children Adolescents | High | Children with ADHD did not display clear problems with basic instrumental learning under laboratory conditions, but may display problems when task complexity increases. Children displayed problems with conditional discrimination learning, a complex form of learning, particularly when there is a delay between the discriminative cue and the opportunity to respond. It is possible that they also display difficulties with reversal learning. No differences were found in regard to the effect of reinforcement on instrumental learning. Significant heterogeneity and some publication bias identified |
Hutchinson | 2008 | MA | 13 P | ADHD n = 284 Controls n = 311 | Adolescents | Mod | Female children and adolescents with ADHD displayed significantly smaller splenium size (moderate effect size). Male children and adolescents with ADHD displayed significantly smaller rostral body size. Participant age did not affect findings. Cannot determine effect of comorbidity. Publication bias unlikely |
Jackson | 2016 | SR MA | 49 N | Participants n = 3913 | Children Adolescents Adults | Mod | Delay discounting is significantly and robustly elevated among children and adults with ADHD (medium effect size). No significant differences based on sample age, reward outcome, or comorbid status was detected. Minimal heterogeneity or evidence of publication bias |
Kaiser | 2020 | SR MA | 53 P | ADHD n = 1576 Control n = 1794 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed on average atypical event rate potential during inhibitory control, attention, working memory, and performance monitoring. On average children and adults with ADHD displayed smaller Cue-P300-amplitudes, longer Go-P300-latencies, smaller NoGo-P300-amplitudes, longer NoGo-P300-latencies, smaller CNV-amplitudes, and smaller Pe-amplitudes. Children with ADHD: larger mean effect sizes for the NoGo-P300-amplitude, Pe-amplitude, Go-P100-latency, Go-P300- latency and NoGo-N200-latency. Adults: largest mean effect sizes for the Cue-P300-amplitude component. Results are characterised by substantial heterogeneity and moderate effect sizes |
Kamradt | 2018 | SR MA | 12 P | Participants n = 1041 | Children Adolescents Adults | Mod | No effect was found between cortisol reactivity and ADHD, although significant heterogeneity may moderate this finding. No evidence of publication bias |
Karalunas | 2014 | SR MA | 18 N | ADHD n = 426 ASD n = 455 Control n = 1065 | Children Adolescents | Mod | Children and adolescents with ADHD displayed significantly slower drift rates (moderate to large effect size), significantly faster nondecision times (small effect size). Findings underlie the reaction time variability differences displayed by individuals with ADHD. No evidence of heterogeneity or publication bias |
Karalunas | 2013 | MA | 7 P | ADHD n = 640 Controls n = 626 | Children Adolescents | Low | Children with ADHD demonstrated more low-frequency reaction variability (small effect size) as well as equivalent excess reaction variability in a faster frequency comparison band (small effect size). There was a trivial and nonsignificant difference between effect sizes in the low and high frequency bands. New data replicated these results. Between study heterogeneity present. Publication bias corrected during meta-analysis |
Kasper | 2012 | SR MA | 45 N | ADHD n = 1989 Controls n = 1874 | Children Adolescents | Mod | Children with ADHD exhibit heterogenous, statistically significant phonological and visuospatial working memory task deficits (large effect size). Several moderative variables (i.e., percentage of female participants, number of experimental trials, recall vs. recognition tasks, and demands on the central executive, explained the significant effect size variability among phonological and visuo-spatial studies. Mild evidence of publication bias within the sample of studies present |
Koenig | 2017 | SR MA | 8 P | ADHD n = 317 Controls n = 270 | Children Adolescents Adults | Mod | Children and adults with ADHD did not display altered short-term measures of resting-state vagal tone. No significant heterogeneity across effect sizes reported. Risk of publication bias not assessed |
Kofler | 2013 | SR MA | 319 N | ADHD n = 9780 Controls n = 12,024 | Children Adolescents Adults | High | Children with ADHD demonstrated robust but heterogenous impairments in reaction time variability relative to controls (large effect size). This increased variability was attenuated by psychostimulant treatment but unaffected by non-stimulant medical and psychosocial interventions. Adults with ADHD continue to demonstrate significantly increased reaction time variability (moderate effect size). Processing speed was unaffected when reaction time variability is accounted for. Publication bias corrected during meta-analysis |
Kowalczyk | 2022 | SR | 34 P | ADHD n = 981 ADHD remit n = 38 Non-ADHD sibling n = 134 Controls n = 774 | Children Adolescents Adults | Mod | Across cognitive domains, children and adults with ADHD consistently displayed atypical functional connectivity in the cingulo-opercular, sensorimotor, visual, subcortical, and executive control networks, and similar sensorimotor and subcortical (primarily striatal) networks. Large heterogeneity in study methodologies prevented a meta-analysis. Low risk of publication bias |
Lanier | 2021 | SR | 11 N | ADHD n = 291 Controls n = 197 | Children Adults | Low | ADHD is commonly associated with impairing spontaneous mind wandering. Risk of publication bias not assessed |
Lansbergen | 2007 | SR MA | 18 N | ADHD n = 757 Controls n = 605 | Children Adolescents Adults | Low | Children and adults with ADHD consistently displayed significantly compromised interference control. Small effect size deemed due to between study methodological differences and an additional random component. Risk of publication bias not assessed |
Lei | 2015 | MA | 23 P | ADHD n = 324 Controls n = 350 | Children Adolescents Adults | Mod | Individuals with ADHD displayed significantly decreased activation during response inhibition in the supplementary motor area, insula, caudate, and precentral gyrus, as well as increased activation in the postcentral gyrus, inferior frontal gyrus, and precuneus. The activation decreases in the R caudate were greater in child ADHD patients than adult ADHD patients. Risk of publication bias not assessed |
LeRoy | 2019 | SR | 16 N | ADHD n = 718 Controls n = 337 | Adults | Low | Adults with ADHD-C and ADHD-I performed worse on measures of executive functioning (tests measured cognitive inhibition, selective attention, cognitive flexibility, processing speed, planning), attention (tests measured orientating attention, response time, vigilance), working memory, and memory. Direct comparisons of the two subtypes showed that those with ADHD-C performed worse in the memory domain. There were few to no differences in other domains. Risk of publication bias not assessed |
Lijffijt | 2005 | MA | 29 N | ADHD n = 977 Controls n = 1078 | Children Adolescents Adults | Low | Children with ADHD but not adults with ADHD display significantly longer basic reaction time (moderate effect size). Adults with ADHD display a slowing of stop sign reaction time relative to mean reaction time (moderative effect size). Risk of publication bias not assessed |
Liu | 2023 | MA | 38 P | ADHD n = 1,126 Controls n = 1277 | Children Adolescents Adults | Mod | Children with ADHD consistently exhibited hyper-connectivity between different parts of the cortex and L middle frontal gyrus, and hypo-connectivity between different parts of the cortex and L putamen/pallidus/ amygdala. Adults with ADHD show hyper-connectivity between the cortex and R precuneus/sub-gyral/cingulate gyrus. Risk of publication bias not assessed |
Loyer-Carbonneau | 2021 | MA | 54 N | ADHD n = 10,212 | Children Adolescents | High | Boys and girls with ADHD presented with significantly more executive (i.e., working memory, planning, cognitive flexibility, motor response inhibition, and interference control) and attentional (i.e., sustained, selective and divided attention) deficits. Boys with ADHD tend to display more hyperactivity as well as more difficulties with response inhibition and cognitive flexibility in comparison to girls with ADHD. Significant heterogeneity between effect sizes noted. Effect sizes did not seem to be affected by a publication bias |
Lukito | 2020 | SR MA | 80 P | ADHD n = 2534 Controls n = 2299 ASD n = 60 | Children Adolescents Adults | Mod | ADHD is associated with reduced grey matter volume and under activation in a number of brain regions during cognitive control and motor response inhibition. Findings somewhat vary between children and adults with ADHD. No evidence of publication bias |
Ma | 2016 | MA | 10 P N | ADHD n = 210 Controls n = 274 | Children Adolescents | Low | Reinforcement can normalise inhibitory control in children and adolescents with ADHD (large effect size). Inhibitory control may also improve to a larger extent in adolescents with ADHD as a function of reinforcement. Potential factors contributing to the reinforcement effects include the reinforcement schedules and the type of reinforcement. Heterogeneity deemed low. Risk of publication bias not assessed |
Marx | 2021 | SR MA | 37 N | Participants n = 3763 (53% with ADHD) | Children Adolescents Adults | Mod | Studies examining single choice paradigm: Individuals with ADHD tended to choose small immediate rewards over large delayed rewards more frequently than controls (small to medium effect size). Significant interstudy heterogeneity and possible publication bias noted. Studies examining temporal discounting: offering real rewards to individuals with ADHD nearly halved choice impulsivity in participants with ADHD. No evidence of heterogeneity and publication bias |
Mauri | 2018 | StR | 11 P | ADHD n = 196 Controls n = 195 | Children Adolescents | Very Low | Children and adolescents with ADHD tended to displayed peculiar cortical activation both during neurological and emotional tasks, and lower (1) prefrontal cortex activation during tasks assessing attention, cognitive inhibition and flexibility, working memory and memory span, and (2) temporal cortex re-activity to facial emotional stimuli. Consistent interpretation of findings is limited due to substantial methodological heterogeneity. Risk of publication bias not assessed |
McCarthy | 2014 | SR MA | 20 P | ADHD n = 334 Controls n = 372 | Adolescents Adults | Mod | Across all tasks, adolescents and adults with ADHD displayed significantly less frontal lobe activity. N-back task: less activity in the bilateral superior frontal gyri and L medial frontal gyrus. Go/no-go tasks: less mean activation in the L medial frontal gyrus and R caudate. Stop tasks: children displayed less activity the bilateral inferior frontal gyri, R superior frontal gyrus and R middle frontal gyri. Risk of publication bias not assessed |
Metin | 2012 | MA | 30 N | ADHD n = 2091 Controls n = 3836 | Children Adolescents Adults | Low | Children and adults with ADHD displayed significant and disproportionate slowing of reaction time on trials with slow event rates. For commission errors, the effect sizes were larger on trials with fast event rates. No event-related effects were seen for reaction time variability. There were also general effects of ADHD on performance for all variables that persisted after effects of event-rate were taken into account. Significant heterogeneity account for by using a random effects model. Risk of publication bias not assessed |
Mette | 2023 | StR | 7 N | ADHD n = 346 Controls n = 286 | Adults | Low | The results of the present review indicate that the number of studies on time perception in adult ADHD is very scarce. Moreover, the main investigated domains of time perception in the past decade were time estimation, time reproduction and time management. Whereas some of the found studies were able to demonstrate a distinct deficit in time estimation, time reproduction and time management other studies were unable to demonstrate a clear association between ADHD and time estimation and time reproduction deficit. Significant clinical and methodological heterogeneity reported. Risk of publication bias not assessed |
Michelini | 2022 | SR MA | 28 P | MA ADHD n = 645 MA Controls n = 521 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed broad brain-oscillatory alterations with small (theta) and small-to-moderate (alpha and beta) effect sizes. These group differences were partly consistent when repeating analyses by age group (< 18 and 18 + years) and task type (cognitive control, working memory, and simple attention tasks). Risk of publication bias not assessed |
Mills | 2018 | LS | P | ADHD n = 256 Controls n = 176 | Adolescents | High | Children with ADHD displayed reduced negative connectivity between task positive (i.e., fronto-parietal, cingulo-opercular, and dorsal attention networks), and task negative networks (i.e., default mode network). Connectivity continues to become more negative throughout development. Children with ADHD displayed poorer signal detection on the continuous performance task (reduced attentional vigilance), more so on easy than difficult tasks |
Mowinckel | 2015 | SR MA | 59 N | Unable to determine | Adults | Mod | Adults with ADHD displayed atypical decision-making. Reward based decision making and continuous performance task: small to medium effect sizes. Drift diffusion model analysis: large effect size. No evidence of publication bias |
Mullane | 2009 | SR | 12 N | ADHD n = 272 Controls n = 280 | Children | Low | Children with ADHD displayed significantly poorer interference control as indexed by reaction time (results were heterogenous), accuracy (large effect size), and inverse efficiency congruency (large effect size). Risk of publication bias not assessed |
Mullane | 2008 | StR | 7 N | ADHD n = 180 Controls n = 193 | Children Adolescents | Low | Children and adolescents with ADHD demonstrated less efficient serial search, both at the lowest and highest levels of task complexity, with difficultly increasing with demand (i.e., increased search complexity and large item numbers). Results were variable for effortful serial search. No differences in automatic search were found. These findings indicate children with ADHD show impairments in aspects of effortful visual selective attention, as measured by visual search. Risk of publication bias not assessed |
Nakao | 2011 | SR MA | 14 P | ADHD n = 378 Controls n = 344 | Children Adolescents Adults | Low | Children and adults with ADHD displayed significantly smaller grey matter volume. This difference was more pronounced when the adult study was excluded. No significant heterogeneity or publication bias noted. Children with ADHD may display significantly and robustly smaller grey matter volumes in the R lentiform nucleus, extending to the caudate, and larger grey matter volumes in the L posterior cingulate cortex. Findings were heterogenous |
Nejati | 2020 | MA | 25 N | ADHD n = 1660 Controls n = 1234 | Children | High | Children with ADHD displayed significant difficulties in time perception. The problem is most pronounced in tasks of longer duration but also obvious during tasks assessing different types of modalities (i.e., visual and auditory), and time estimation, reproduction and anticipation. Heterogeneity in the published studies was large and statistically significant. No evidence of publication bias |
Onandia-Hinchado | 2021 | SR | 93 N | ADHD n = 5574 Controls n = 4880 Other n = 1323 | Adults | Low | Adults with ADHD displayed atypical attention regulation (alertness, selective and sustained), processing speed, executive function (mainly working memory and inhibition with emphasis on reward delay and interference control), verbal memory, reading skills, social cognition and arithmetic abilities. Risk of publication bias not assessed |
Oosterlaan | 1998 | MA | 8 N | ADHD n = 180 CD/ODD n = 68 ADHD/CD/ODD n = 65 Controls n = 159 Other n = 74 | Children | Low | Children with ADHD displayed poor response inhibition related to a slow inhibitory process during tasks measuring Go Response Mean Reaction Time (medium effect size), Inhibition Function Slope (large effect size), and Stop Signal Reaction Time (medium effect size). The findings for children with conduct disorder were similar but less consistent. Children with ADHD and conduct disorder display greater impairments than children with only ADHD. Risk of publication bias not assessed |
Orban | 2022 | MA | 20 N | ADHD n = 641 Controls n = 557 | Children Adolescents Adults | High | Across all studies examining constructs of memory control, individuals with ADHD demonstrated worse control over memory and differences, along with differences in verbal memory interference control during the following measurement tasks: continuous moderators of practice interference (children only, moderate effect size, heterogeneity corrected during meta-analysis) and retroactive interference (children and adults, non-significant heterogeneity). Homogeneity was significant. No evidence of publication bias |
Paloyelis | 2007 | SR | 20 P | Unable to determine | Children Adolescents Adults | Very Low | Individuals with ADHD display both lower and higher brain activity in different regions of the brain dependent on task. In analyses that examined inhibition errors, as well as in tasks that tapped attention processes, motor function and working memory, the ADHD group almost exclusively showed lower activity. In the attentional tasks, this was mostly over temporal and parietal areas; in motor function tasks, mostly over frontal areas. Risk of publication bias not assessed |
Patros | 2019 | SR MA | 41 N | ADHD n = 2051 Controls n = 2766 | Children | High | Children with ADHD experienced significant atypical planning (moderate effect size) and responded more quickly on planning tasks (small to moderate effect size). Participant age, percentage of females, solution presentation (e.g., pictorial vs. physical display), and task complexity (beads vs. disks) were identified as statistically significant moderating variables across planning metrics. No evidence of publication bias noted in the results pertaining to planning accuracy, composite planning performance, rule violations, problem-solving time and initiation time when baseline motor-response time controlled. The potential for publication bias was noted in regard to the initiation time and solution time results |
Patros | 2016 | SR MA | 26 N | ADHD n = 2360 Controls n = 1960 | Children Adolescents | High | Children and adolescents with ADHD exhibited increased impulsive decision-making (moderate effect size). When choice-impulsivity tasks were examined independently, children with ADHD performed significantly more impulsively on both delay of gratification and delay discounting tasks. The use of single-informant diagnostic procedures relative to multiple informants yielded larger between-group effects, and a similar pattern was observed across samples that excluded females relative to samples that included females. No evidence of publication bias noted in the results pertaining to choice impulsivity and delay of gratification. The potential for publication bias was noted in regard to the delay discounting results |
Pauli-Pott | 2015 | MA | 19 N | ADHD n = 1395 Controls n = 1195 | Children Adolescents | Mod | The executive inhibitory control and delay aversion/discounting challenges children with ADHD experience during complex cool inhibition tasks (which specifically involve the executive attention system) increased with age between 3 and 6 years (moderate effect size), but appear to become smaller, although still significant (small effect size) when studies include adolescents aged between 13 and 16 years. There was no significant difference between studies that used versions of the delay/temporal discounting task vs. other tasks. Results were not influenced by proportion of boys and co-morbidity. Risk of publication bias not assessed |
Pievsky | 2018 | SR of MA | 34 MA N | Unable to determine | Children Adolescents Adults | Mod | ADHD is associated with heterogenous difficulties across a variety of neurocognitive including set shifting, working memory reaction time variability response inhibition, vigilance, intelligence/achievement and planning/organisation (all moderate effect size except for set shifting which had a small effect size). Risk of publication bias corrected during meta-analysis. Age moderated the relationship between ADHD diagnosis and neurocognitive functioning, with greater between-groups differences among children and adults than among adolescents. Studies that received drug funding reported larger effect sizes than those without |
Plichta | 2014 | MA | 10 P | ADHD n = 231 Controls n = 185 | Children Adolescents Adults | Very low | Individuals with ADHD displayed ventral-striatal hypo-responsiveness during task measuring reward anticipation (moderate effect size). The included studies were homogenous. Risk of publication bias not accessed |
Posner | 2014 | SR | 24 P | ADHD n = 583 Controls n = 927 | Children Adolescents Adults | Very Low | ADHD is associated with heterogenous anticorrelations (1) between the cognitive control network and the default mode network, (2) within the default node network and the cortioco-striato-thalamo-cortical loops (cognitive and limbic). Risk of publication bias not assessed |
Proal | 2011 | LS | - | ADHD n = 207 Mean age at entry 8 years Controls n = 178 | Children Adolescents | High | Adults diagnosed with childhood ADHD displayed significantly (1) reduced cortical thickness in the dorsal attentional network and limbic areas (emotional regulation and motivation), and (2) decreased grey matter in the R caudate, R thalamus and bilateral cerebellar hemispheres (R hemisphere clusters were located in the inferior parietal lobe, temporal pole, and insula; L hemisphere clusters were located in superior frontal gyrus/frontal pole, precentral gyrus, insula, temporal pole, and cuneus). They also displayed cortical thinning in the bilateral parietal lobes, occipital lobe, temporal poles, insula, precentral gyri, frontal poles, and R precuneus |
Ramos | 2020 | SR MA | 49 N | ADHD n = 4956 Controls n = 3249 | Children Adolescents | Mod | Children and adolescents with ADHD exhibited robust poorer working memory performance during digital scan backwards performance (medium effect size). Effect size reduced with increasing age and explained a moderate level of heterogeneity, with younger children presenting with the greatest difficulties. A small level of publication bias noted |
Rash | 2012 | SR | 6 P | ADHD n = 155 Controls n = 89 | Children Adolescents | Low | Findings were not unanimous but suggest children with unmedicated ADHD experienced lower levels of cardiac vagal tone during tasks that involved self-regulation and emotion regulation. Medication acted to correct the autonomic imbalance experienced by children with ADHD but did not bring this imbalance into normal levels. Risk of publication bias not assessed |
Robe | 2019 | SR MA | 13 P | ADHD n = 869 Controls n = 909 Children, adolescents | Children Adolescents Adults | High | Individuals with ADHD displayed heterogenous reduced vagally-mediated heart rate variability post task demand (small affect size). Effect size was greatest in the presence of co-morbidity. Task type and respiration rate also significantly moderated findings. Significant risk of publication bias noted |
Roberts | 2021 | MA | 116 N | Unable to determine | Children Adolescents Adults | Mod | Children and adults with ADHD exhibit more risk-taking compared to children and adults without the disorder (moderate effect size). Larger effect sizes were associated with studies involving virtual reality simulations and those in which risky options were disadvantageous, compared to studies in which risky options were more advantageous or similar to safe alternatives. No evidence of publication bias was noted for results related to self-reported measures and virtual reality tasks. For studies encompassing behaviour tasks publication bias may be present |
Samea | 2019 | SR MA | 96 P | n = 1914 | Children Adolescents | Mod | Pooled structural and functional, sub-analyses restricted to modality, and in-/decreased contrast did not yield any significant finding. Sub-analysis of task t-fMRI experiments using neutral stimuli resulted in convergence of aberrant ADHD-related finding in the L pallidum/putamen. Decreased activity in the L inferior frontal gyrus was also found in male subjects only. Lack of regional convergence in other areas in children/adolescents with ADHD may be due to heterogeneous clinical populations, various experimental design, pre-processing, and statistical procedures in individual publications. Risk of publication bias not statistically assessed |
Sanjeevan | 2020 | SR MA | 7 N | ADHD n = 213 Controls n = 257 | Children Adolescents Adults | High | Procedural sequence learning appears to be preserved in ADHD. Heterogeneity was significant across studies and could be partially attributed to the age of participants. No evidence of publication bias |
Schoechlin | 2005 | SR MA | 24 N | ADHD n = 867 Controls n = 806 | Adolescents Adults | Mod | In all functional domains, adults with ADHD scored significantly lower in the domains of visual and verbal fluence, abstract problem-solving, sustained attention, focused attention, verbal memory (all moderate effect size), as well as verbal intelligence (reasoning), executive functions, visual/figural problem-solving, simple attention and figural (visual) memory (all small effect size). Heterogeneity was only significant for visual/verbal fluency and simple attention. Risk of publication bias deemed unlikely |
Schwartz | 2008 | MA | 25 N | ADHD n = 1535 Controls n = 1730 | Children Adults Adolescents | Mod | Response inhibition measured using the Stroop interference effect (the tendency to experience difficulty naming a physical colour when it is used to spell the name of a different colour) was not larger in children or adults with ADHD regardless of age. Children and adults with ADHD however responded on average 1.14 times slower during both the colour and the colour-word condition. Risk of publication bias not assessed |
Senkowski | 2022 | SR MA | 26 N | ADHD n = 883 Controls n = 916 | Adults | High | Adults with ADHD display reliable inhibitory control deficits, as expressed in prolonged stop signal task response times (moderate effect size). Results were not moderated by study quality, sample characteristics or clinical parameters. Adults with ADHD may also display greater stop signal task omission errors and reduced go accuracy (small effect size), indicative of altered sustained attention. Subtle publication bias may be present |
Shaw | 2013 | LS | P | ADHD n = 92 Controls n = 184 | Adolescents Adults | High | Adult ADHD status is associated with altered development and increased cortical thinning of components of the cortical networks and support attention, cognitive control, and the default mode network, Specifically, the R cingulate gyrus (extending to the R caudal regions and R isthmus, and extending to the R precuneus of the superior parietal lobe, R cuneus of the occipital lobe, and R fusiform gyrus), R dorso-lateral prefrontal cortex, L caudal anterior cingulate, L para-central lobule, L precuneus, and L postcentral gyrus |
Shaw | 2012 | LS | P | ADHD n = 234 Controls n = 231 | Children | High | Although childhood maturation of the cortical surface area progressed along the typical trajectory, surface area development was delayed in children with ADHD. When children first participated in the study (mean age 10.4 years), cortical surface reduction was most pronounced bilaterally in the prefrontal cortex (especially the medial wall and the lateral superior, middle, and polar frontal regions), the R lateral temporal cortex, and L medial temporal cortex, extending to posterior L medial wall (fusiform, lingual gyri, and cuneus). At a sub-lobular level, delay was most prominent in the anterolateral prefrontal gyri bilaterally, especially on the right. Additionally, delay was prominent bilaterally in the medial temporal gyri, in the R postcentral and middle temporal gyri, and L supramarginal gyrus. For the R prefrontal cortex, the median age by which children with ADHD achieved attained peek area in 50% of cortical vertices was 14.6 years, significantly later than controls at 12.7 years. A similar, but less pronounced, delay was found in the L hemispheric lobes. There were no such diagnostic differences in the developmental trajectories of cortical gyrification |
Shaw | 2007 | LS | P | ADHD n = 223 Controls n = 223 | Children | High | Although regional brain maturation progresses in a typical manner, children with ADHD displayed marked delay in attaining peak thickness throughout most of the cerebrum. The mean age at which children with ADHD attained peak thickness for 50% of the cortical points was 10.5 years, which is significantly later than the mean age of 7.5 years for controls. Differences were most prominent in the middle prefrontal cortex, where the ADHD group reached their peak thickness approximately 5 years after controls, and to a lesser extent in the superior prefrontal and medial prefrontal cortex. Delay was also found in the temporal cortex extending to the middle occipital gyri, most prominently in the posterior portions of the middle and superior temporal gyrus bilaterally |
Shaw | 2006 | LS | P | ADHD n = 163 Controls n = 166 | Children | High | Children with ADHD displayed global cortical thinning, most prominently in the medial and superior prefrontal and precentral regions, including the superior and medial frontal gyri and cingulate region bi-laterally, L precentral gyrus, and R anterior/mesial temporal cortex. Cortical thickness developmental trajectories did not differ significantly between participants with ADHD and control groups throughout except in the R parietal cortex, where trajectories converged by age 17 |
Skodzik | 2017 | SR MA | 24 N | ADHD n = 827 Controls n = 771 | Adolescents Adults | Mod | Adults with ADHD displayed significant verbal working memory deficits (medium effect size) but not visual working memory deficits during tests examining delayed free recall, memory acquisition and recognition memory. Results were heterogeneous. No evidence of publication bias. The long-term memory performance problems adults with ADHD experience are strongly influenced by deficits already present in the stage of memory acquisition and reflects a learning (memory acquisition) challenge induced at the stage of encoding |
Snyder | 2006 | MA | 9 P | n = 1498 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed a consistent increase in theta/beta ratio (medium effect size). Heterogeneity was attributed to (1) meta-analytic technique, (2) eyes open versus eyes closed, and (3) control of medication. Risk of publication bias not assessed |
Sutcubasi | 2020 | SR MA | 20 P | ADHD n = 944 Controls n = 1121 | Children Adolescents Adults | Mod | Children and adults displayed within-default mode network hypo- and hyper-connectivity, and within-cognitive control system hyperconnectivity. Children and adolescents also displayed reduced connectivity between the default mode and cognitive control networks, and the affective/motivational and salience networks. Risk of publication bias not assessed |
Szuromi | 2011 | SR MA | 6 P | ADHD n = 154 Controls n = 140 | Adolescents Adults | Low | Adults with ADHD displayed significantly reduced P3 amplitude in comparison to controls (medium effect size). The effect size is greater in females than males, and the reduction in P3 amplitude appears to increase with age. Risk of publication bias not assessed |
Talbot | 2018 | SR | 6 N | ADHD n = 162 Controls n = 176 | Children Adolescents | Low | Children with ADHD performed significantly worse on time-based prospective memory tasks and event-based prospective memory tasks (both considered a type of long-term memory or involving long-term memory), irrespective of their ongoing task performance. Risk of publication bias not assessed |
Valera | 2007 | MA | 21 P | ADHD n = 565 Controls n = 583 | Children Adolescents | Mod | Children with ADHD displayed heterogenous global volume reductions, the largest differences include cerebellar regions, the splenium of the corpus callosum, total and R cerebral volume, and R caudate. Risk of publication bias not statistically assessed |
Valmiki | 2021 | SR | 9 N | ADHD n = 493 Controls n = 330 | Adolescents (male only) | Mod | Male adolescents with ADHD are more responsive to reinforcement related to monetary gain and loss and are more sensitive to larger rewards than smaller ones. ADHD-C and typically developed controls were found to have similar responsiveness, while ADHD-I appeared to be insensitive to stimulus. Risk of publication bias not assessed |
Van Ewijk | 2012 | SR MA | 15 P | ADHD n = 173 Controls n = 169 | Children Adolescents Adults | Mod | Children and adults with ADHD displayed wide spread alterations in white matter integrity, most evidently in the R anterior corona radiata, R forceps minor, bilateral internal capsule, and L cerebellum. Other areas that were less reliably implicated include the inferior and superior longitudinal fasciculus, cingulum, corpus collosum, caudate, basal ganglia, uncinate fasciculus, forceps minor, frontal, temporal, parietal and occipital lobes. Risk of publication bias not assessed |
Vidor | 2022 | SR MA | 33 P | ADHD n = 874 | Children | High | Children with ADHD displayed higher concentrations of a composite of glutamate and glutamine in the R medial frontal area. Glutamate may be implicated in pruning and neurodegenerative processes as an excitotoxin, while glutamine excess might signal a glutamate depletion that could hinder neurotrophic activity. Both neuro metabolites could be implicated in the differential cortical thinning observed in patients with ADHD across all ages. Risk of publication bias not assessed |
Vieira de Melo | 2018 | SR | 6 P | n = 4144 | Children Adolescents Adults | Low | Childhood ADHD is associated with overall and region-specific cerebral volume reductions, and hypo- and hyper-activation. Adults with ADHD displayed region-specific frontal lobe hypoactivation and parietal lobe hyperactivation. It was impossible to determine if age, gender, and comorbidity accounted for heterogeneity. Risk of publication bias not assessed |
Willcutt | 2005 | MA | 83 N | ADHD n = 3734 Controls n = 2969 | Children Adolescents Adults | Low | Children and adults with ADHD displayed significant weaknesses on all executive function measurements (medium effect size). The strongest and most consistent effects were obtained on measures of response inhibition, vigilance, working memory, and planning. Adjusting for heterogeneity did not influence effect size. Effects could not be attributed to variables such as intelligence, reading ability, and symptoms of other disorders. Risk of publication bias not assessed |
Yap | 2021 | SR | 31 P N | ADHD n = 667 Controls n = 698 | Children Adolescents Adults | Mod | Individuals with ADHD displayed activation alterations in the frontal and parietal regions and poorer sustained attention and working memory, regardless of age and task paradigm. Functional heterogeneity could be attributed to different brain activation patterns, large intra-subject variability, or both. Risk of publication bias not assessed |
Zhang | 2023 | MA | 47 P | ADHD n = 1150 ASD n = 648 Controls n = 1615 | Children Adolescents Adults | High | Children with ADHD do not display significant fractional anisotropy changes. Adolescents with ADHD displayed decrease fractional anisotropy in the corpus callosum (body and splenium), L optic radiations, R superior longitudinal fasciculus, L pons, inferior longitudinal fasciculus, and R corticospinal projections. Adults with ADHD displayed increased fractional anisotropy in the L superior longitudinal fasciculus, body of the corpus callosum and R precuneus as well as decreased fractional anisotrophy clusters in the genu, body and splenium of the corpus callosum |
Zhang | 2018 | SR MA | 10 P | ADHD n = 654 Control n = 529 | Children Adolescents | High | Only males with ADHD displayed significantly higher peripheral brain-derived neurotrophic factor levels (a key molecule involved in differentiation of neuronal populations during development and plastic changes related to learning and memory). High heterogeneity was noted across sampled studies, which may be a function of sample size, participants sampled, variations in study design, or other factors. Risk of publication bias not assessed |
Zhao | 2022 | MA | 66 P | n = 4137 | Children Adolescents Adults | High | Children and adults with ADHD displayed decreased fractional anisotropy. Decreased fractional anisotropy was identified in one cluster in the corpus callosum via anisotropic effect size-signed differential mapping. Results may be affected by potential publication bias. Decreased fractional anisotrophy was identified at the genu, splenium of the corpus callosum and parietal lobe via activation likelihood estimation. No evidence of publication bias. Findings indicate ADHD is associated with the presence of aberrant white matter microstructure |
Zheng | 2022 | MA | 27 N | ADHD n = 1620 Controls n = 1249 | Children Adolescents | High | Children and adolescents with ADHD perceived time less accurately (medium effect size) and less precisely (medium effect size) and displayed a higher tendency to overestimate time than controls. Moderator analyses indicated that the discrepancy of time perception between groups was not affected by the type of timing tasks nor the modality of stimuli used in the tasks, but was influenced by participant age with time perception abilities improving in terms of both accuracy and precision, with increased age. Low risk of publication bias for all results reported |