Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews

journal homepage: www.elsevier.com/locate/neubiorev

Is autonomic nervous system function atypical in attention deficit T

hyperactivity disorder (ADHD)? A systematic review of the evidence
Alessio Bellatoa,*, Iti Aroraa, Chris Hollisa,b, Madeleine J. Grooma
a
Division of Psychiatry and Applied Psychology, School of Medicine, University of Nottingham, Institute of Mental Health, Innovation Park, Triumph Road, Nottingham,
NG7 2TU, United Kingdom
b
NIHR MindTech Healthcare Technology Co-operative, Institute of Mental Health, Innovation Park, Triumph Road, Nottingham, NG7 2TU, United Kingdom

ARTICLE INFO ABSTRACT

Keywords: Although arousal mechanisms have frequently been found to be atypical in ADHD, these findings usually
ADHD emerged from indirect behavioural measures which give only a limited understanding of arousal dysregulation
Arousal regulation in this condition. To assess the hypothesis that functioning of the autonomic nervous system (ANS), one com-
Autonomic nervous system ponent of arousal, is atypical in ADHD, we carried out a systematic review of the literature on 55 studies
Heart rate
investigating electro-dermal, heart rate and pupillometry measures under different experimental conditions
Electrodermal activity
Skin conductance
(resting-state, cognitive tasks and in response to reinforcers or socio-emotional stimuli). Our literature review
Pupillometry identified ANS dysfunction in individuals with ADHD, more often in the direction of hypo-arousal than hyper-
Resting-state arousal, particularly at rest and during tasks requiring response regulation and sustained attention. Almost half
Cognitive of the reported findings were null. Stimulant medications increased ANS activity and, in some studies, re-
Socio-emotional inforcers and rewards produced a similar effect, suggesting that ANS function can be modified in ADHD. Further
Rewards research is needed to assess the influence of comorbid symptoms and to explore methodological parameters that
Stimulant medication may influence findings.

1. Introduction condition (Kuntsi and Klein, 2012; Sergeant, 2000; Van Der Meere,
2002; Van Der Meere et al., 2010). Specifically, manipulations such as
Attention deficit/hyperactivity disorder (ADHD) is a common and optimising the event rate within a cognitive task or providing perfor-
enduring neurodevelopmental disorder characterised by devel- mance-based incentives may improve cognition in ADHD by stimu-
opmentally atypical inattention and/or hyperactivity and impulsiveness lating arousal and thereby reducing the effort required to complete a
(American Psychiatric Association, 2013). The condition affects around cognitive task (Sergeant, 2000; 2005). To date, however, much of the
5 % of children (Polanczyk et al., 2014) and 3 % of adults (Fayyad et al., research in this area has inferred impaired arousal regulation in ADHD
2007) worldwide, resulting in lifelong impairments in most cases, in- from cognitive performance measures, such as reduced accuracy,
cluding mental health problems, unemployment and criminality slower response speed and increased reaction time variability (RTV)
(Erskine et al., 2016). Compared to typically developing controls, in- (Karalunas et al., 2014). Indirect measures, such as these, give only a
dividuals with ADHD are cognitively impaired (see Frazier et al., 2004, limited understanding of arousal dysregulation in ADHD.
for a meta-analysis), with impairments most commonly found in sus- Arousal refers to the neural, behavioural and physiological me-
tained and selective attention (Mueller et al., 2017) and executive chanisms that regulate states of wakefulness and alertness, which are
functions (EF) (Willcutt et al., 2005). However, when cognitive tasks governed by interactions between the peripheral and central nervous
require less effort (Borger and van der Meere, 2000), when stimulus systems (CNS). The autonomic nervous system (ANS) forms one part of
event rate is optimal (Wiersema et al., 2006; 2014) or when rewards are the peripheral nervous system and regulates bodily functions (including
given for performance (Groom et al., 2010, 2013; Liddle et al., 2011), heart rate, respiration, perspiration and pupil dilation) by controlling
children and adults with ADHD exhibit fewer cognitive impairments. To smooth muscle fibres, cardiac muscle fibres and glands. The two
account for these features, it has been suggested that a decreased ability branches of the ANS, the Sympathetic (SNS) and Parasympathetic
to regulate arousal may contribute to the higher-level cognitive deficits Nervous Systems (PNS), exert opposing forces on one another to facil-
in ADHD, and this may be an important aspect of the pathology of the itate constant and dynamic shifts in ANS activity, depending on the

⁎
Corresponding author.
E-mail address: [email protected] (A. Bellato).

https://doi.org/10.1016/j.neubiorev.2019.11.001
Received 16 May 2019; Received in revised form 31 October 2019; Accepted 1 November 2019
Available online 10 November 2019
0149-7634/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

requirements of a given environment or task. The ANS is, therefore, a (vmPFC)), insula, hypothalamus and amygdala (see Fig. 1). The LC is
core component of the arousal system, but its role in the clinical and the sole source of norepinephrine (NE) in the cortex and NE availability
cognitive features of ADHD is not well understood. To advance in these regions influences a range of cognitive functions, including
knowledge of ANS function in ADHD, we conducted a systematic re- perception, memory, working memory, sustained attention and task
view of studies investigating ANS activity in ADHD in the context of a switching (Sara and Bouret, 2012), partly by modulating the avail-
cognitive task or a defined resting-state period. Specifically, we pre- ability of dopamine (DA) and glutamate at task-relevant sites (Mather
dicted that if arousal dysregulation is a feature of ADHD, as suggested et al., 2016).
by previous research cited above, and if this is at least partly due to Animal (Aston-Jones and Cohen, 2005) and human (Gilzenrat et al.,
dysfunction in the ANS, measures of ANS activity will differ sig- 2010; Murphy et al., 2014) studies have demonstrated the concurrent
nificantly between individuals with ADHD and typical control partici- involvement of ANS, brainstem, and cortical systems in the dynamic
pants either at rest (when not engaged in a specific activity) and/or regulation of behaviour and cognition. More specifically, in humans the
during a task. Reviewing the literature on ANS activity in ADHD will, correlations between activity in BOLD signals in ACC, vmPFC, and in-
therefore, provide useful information to help guide theories of arousal dices of ANS functioning such as heart rate (Critchley et al., 2003;
regulation in ADHD and may also prove useful in understanding how Matthews et al., 2004) and electrodermal activity (EDA; Critchley et al.,
medications and other therapies exert their effects. Before presenting 2003; Nagai et al., 2004; Zhang et al., 2014) reflect the ongoing and
the methods and results of the review, we first describe relationships dynamic integration of information about task demands (represented in
between ANS activity and cognition and, briefly, how these relation- cortex), the physiology required to meet those demands (signals be-
ships might be impaired in ADHD. tween the ANS and brainstem) and the regulation of arousal to support
specific behaviours (top-down control by cortical systems over LC).
1.1. Relationships between the autonomic nervous system (ANS) and While in animals the direct measurement of ANS functioning is possible
cognition and it is widely used, investigating arousal in humans can be achieved
by analysing peripheral indices of autonomic arousal, e.g., heart rate,
The well-known Yerkes-Dodson law (Yerkes and Dodson, 1908) pupil size and EDA (Wass et al., 2015).
describes an inverted U-shaped relationship between arousal and cog- While heart rate (HR) indicates the average number of beats per
nitive performance, with task-directed behaviour being negatively af- minute (BPM), heart rate variability (HRV) represents a measure of the
fected by too low or too high levels of arousal and requiring optimal fluctuations in heart rate over time. HRV has been shown to reflect the
regulation to achieve good performance. These links between arousal parallel activation of the SNS and the PNS. Specifically, acceleration of
and cognition are governed by interactions between CNS and ANS heart rate mirrors the activation of the SNS, while activity in the PNS is
(Aston-Jones et al., 2000; Aston-Jones and Cohen, 2005). In particular, associated with heart rate decelerations (Wass et al., 2015). Moreover,
the locus coeruleus (LC) in the brainstem pons region receives auto- studies have shown that activity in the LC-NE system is accompanied by
nomic signals via the nucleus tractus solitarius (NTS) (Critchley and excitatory effects on cardiac muscles resulting in activation of the SNS
Garfinkel, 2018) and has widespread, reciprocal connections with pre- (Wang et al., 2014) and deactivation of the PNS (Samuels and Szabadi,
frontal cortex (PFC) regions (including anterior cingulate cortex (ACC), 2008), demonstrating a direct relationship between HR and LC activity.
orbitofrontal cortex (OFC) and ventromedial prefrontal cortex Similarly, the constriction and dilation of the pupil are influenced

Fig. 1. Visual representation of the LC-NE system in the human brain (created on https://biorender.com/).

183
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

by activity in the SNS and the PNS (Bast et al., 2018) and studies have 2012). This has been interpreted as a potential marker of impaired
shown a direct correlation between slow fluctuations in LC activity and arousal regulation in ADHD since it is thought to reflect fluctuations in
pupil size (Rajkowski, 1993; Murphy et al., 2014). Therefore, mea- performance due to difficulties in maintaining an optimal level of vig-
suring pupil size during resting-state or cognitive tasks is likely to give ilance (Borger and van der Meere, 2000; Sergeant, 2005).
an indication of ANS activity (Bast et al., 2018). Finally, electrodermal Although informative, these studies provide only an indirect as-
activity (EDA) is a measurement of changes in the constriction and sessment of arousal in ADHD because they are based on performance
dilation of blood vessels underneath the surface of the skin. Changes in measures. The brainstem regions that mediate the link between ANS
skin conductance level (SCL) reflect changes in these veins and thus and the neural systems that support cognition and behaviour, such as
reflect changes in the activity of the ANS (Wass et al., 2015). Together, the LC, have not been studied thoroughly in ADHD. Neither there is a
these indices of ANS activity represent useful measures of changes in clear picture of the degree of dysfunction of the ANS which is likely to
autonomic arousal over time and under specific conditions. play a crucial role in arousal regulation. It is therefore unclear which
mechanisms or systems supporting the interface between autonomic
1.2. The relation between autonomic arousal, cognition and behaviour in arousal and cognition are affected in ADHD. A deeper understanding of
ADHD the functioning of the ANS in ADHD may increase our knowledge about
the mechanisms underpinning this neurodevelopmental condition and
Arousal dysregulation has frequently been described as an im- facilitate new therapies which would target the symptoms more effec-
portant feature of ADHD. Specifically, difficulties regulating arousal tively.
according to situational demands may contribute to the behavioural
phenotype found in ADHD. For example, being unable to increase 2. Review aims and methods
arousal during a monotonous or challenging cognitive task might un-
dermine the type of attentive behaviour required to complete school- To assess the evidence of ANS dysfunction in ADHD, we conducted a
work. Similarly, an impairment in the ability to dampen heightened systematic review of the literature, with the aim of identifying articles
arousal may result in maladaptive behavioural strategies of arousal comparing ANS activity at rest or during a cognitive task between
regulation, such as shouting or running. Geissler et al. (2014) suggest children, adolescents and adults with ADHD and typical individuals.
that reduced vigilance and allocation of attentional resources to the The review aimed to answer the following questions: 1) Is there evi-
environment may be core to ADHD, reflecting a tonically hypo-aroused dence of atypical ANS function in ADHD? 2) If so, does this take the
state, while hyperactivity and impulsive behaviours may be a con- form of hypo-arousal or hyper-arousal? 3) And is it only in the resting
sequent autoregulatory strategy to enhance arousal by creating a sti- state (indicating deficient tonic, baseline arousal) or is it also evident in
mulating environment and so stabilize vigilance. Other features of response to a cognitive and emotional stimulus of some sort (indicating
ADHD which are suggestive of a disturbance in arousal regulation in- deficient regulation of phasic arousal)?
clude emotional dysregulation (Faraone et al., 2019), sleep disorders We performed searches of PsycInfo, MEDLINE and EMBASE data-
(Hvolby, 2015), dysregulation of the Hypothalamus-Pituitary-Adrenal bases from 1975 to 18th December 2018 using keywords in the fields of
(HPA) axis (Isaksson et al., 2012) and problems regulating appetite ADHD, attention, autonomic nervous system, arousal and arousal reg-
(Hanc and Cortese, 2018). ulation (see Fig. 2 for a PRISMA flowchart of the articles screened, from
It is noteworthy that medications for ADHD operate by altering the Moher et al., 2009). These terms were supplemented with words that
availability of NE and DA at cortical sites (Faraone and Buitelaar, describe the key measures used to assess ANS function, including pupil
2010), and often have unwanted autonomic-related side effects, such as dilation, heart rate, heart rate variability and electrodermal activity/
increases of heart rate and blood pressure, suggesting that part of their galvanic skin response. Meta-analysis was not performed as there were
mechanism of action may be to alter ANS activity. Whether or not these too few similarities between study methods and measures. The full
ANS effects contribute to the clinical efficacy of these medications, search strategy, including details of inclusion/exclusion criteria, are
needs to be established. Cognitive performance of children and adults available in supplementary materials (SM1). Full-text articles were
with ADHD can also be improved following exercise interventions (Ng obtained for all those retained and were again reviewed against inclu-
et al., 2017) and neurofeedback (Arns et al., 2014), both of which may sion/exclusion criteria, before extracting data on key features of each
influence arousal. This offers some promise for the development of new included article. A discussion was held between reviewers involved in
therapies for ADHD which, by targeting arousal, may improve the the screening process to reach a decision for any articles that were
cognitive and behavioural features of the condition. To develop or re- unclear. Finally, papers were broadly grouped into those presenting
fine interventions, however, a much fuller understanding of arousal in evidence of hyperarousal or hypoarousal in the patient group compared
ADHD is needed. with a control group, or no group differences.
The application of the cognitive energetic model (Sanders, 1983) to
ADHD by Sergeant (2000; 2005) was the first notable attempt to link 3. Results
together cognition and arousal in ADHD, by describing the roles of
effort (described as the energy needed to meet task demands), activa- Fifty-five studies were included after full-text review (see Fig. 2). A
tion (conceptualised as tonic arousal) and energy (likened to phasic, summary of the ANS measures used (including their abbreviations and
stimulus-locked arousal) in supporting cognitive function. Although acronyms), alongside a description of each measure and their re-
these components of the model have proven difficult to test empirically lationship with ANS functioning, can be found in Table 1. Thirty-two
(Johnson et al., 2009), empirical studies designed to test more broadly studies reported data from electro-dermal activity (EDA) measures, ei-
the concept of state (dys)regulation have supported a potential role for ther as the sole measure (n = 19) or in combination with others
arousal in ADHD (Strauß et al., 2018). In particular, evidence that (n = 13); 35 reported data from heart rate (HR) (22: HR only; 13: HR
cognitive functions improve in ADHD when tasks are combined with alongside other measures); 4 studies reported measures of pupil dilation
performance-based incentives (Liddle et al., 2011; Groom et al., 2010, and only one of these measured pupil dilation in combination with
2013) or are delivered at an optimal pace (Metin et al., 2012; Wiersema other measures. In total, there were 91 findings (i.e. measurements of
et al., 2006), suggest that these factors help to offset impaired arousal group effects) from these 55 studies.
regulation in ADHD. Further support for this comes from evidence of
increased intra-individual reaction time variability (RTV) across a 3.1. Resting state
range of experimental paradigms (see Kofler et al., 2013, for a meta-
analysis), particularly during monotonous cognitive tasks (Metin et al., Resting-state refers to a defined period of time when participants

184
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Fig. 2. PRISMA flowchart describing the numbers of studies identified, screened, excluded and included in the systematic review process.

are not performing any task or activity. Atypical modulation of resting- autonomic arousal. It is therefore possible that these finding of hyper-
state activity has been found in individuals with ADHD, who show in- arousal reflect the way EDA measures were collected or analysed.
creased power in slow- relative to fast-oscillations in EEG (Barry et al., Among the five studies which reported no significant group differences,
2003) and atypical activation of resting-state networks, such as the Herpertz et al. (2003); Iaboni et al. (1997); McQuade and Breaux
Default-Mode Network (DMN) (Rubia, 2018). It is not clear, however, (2017), and McQuade et al. (2017) found no group effect when ana-
whether these atypicalities are accompanied by ANS dysfunction. lysing mean SCL, while Hermens et al. (2005b) did not find any dif-
ference when calculating the slope of SCL throughout the entire resting
3.1.1. Electrodermal activity (EDA) period.
Reduced EDA has been reported by nine studies (out of 16) during
resting-state periods (Table 2) in children/adolescents (Barry et al., 3.1.2. Heart rate
2009, 2012; Beauchaine et al., 2001; Clarke et al., 2013; Crowell et al., Of the twenty studies that analysed heart rate (HR) during resting
2006; Dupuy et al., 2014; Herpertz et al., 2001; Lazzaro et al., 1999) (Table 3), the majority (n = 12) found no group differences on this
and adults with ADHD (Hermens et al., 2004), while two reported measure, while six reported hypo-arousal in ADHD (Beauchaine et al.,
hyper-arousal (Hermens et al., 2005a; Mukhopadhyay et al., 1997). 2001; Crowell et al., 2006; de Carvalho et al., 2014; Griffiths et al.,
Barry et al. (2009; 2012) found signs of hypo-arousal during resting 2017; Herpertz et al., 2003; Wang et al., 2013) and two reported evi-
in ADHD, reflected in reduced mean skin conductance level (SCL; see dence of hyper-arousal (Leikauf et al., 2017; Rukmani et al., 2016).
Table 1 for a description of EDA measures and abbreviations), similarly Some studies suggested that hypo-arousal at rest, in ADHD, may be
to Clarke et al. (2013); Dupuy et al. (2014); Hermens et al. (2004) and mainly associated with increased activation of the PNS. For example,
Lazzaro et al. (1999). Using non-specific skin conductance responses Wang et al. (2013) found that typically-developing male pre-schoolers
(NS-SCRs), Beauchaine et al. (2001) and Lazzaro et al. (1999) found with more inattentive/hyperactive traits showed lower sympathetic and
signs of hypo-arousal in children and adolescents with ADHD (reduced higher parasympathetic activity, by analysing Heart Rate Variability
NS-SCRs), as did Crowell et al. (2006). However, in this last study, it (HRV) frequency measures. Similarly, de Carvalho et al. (2014) found
was not possible to determine whether these effects were driven by signs of hyper-activation of the PNS branch in ADHD (i.e., increased
symptoms of ADHD or symptoms of oppositional defiant disorder NN50 and increased Poincarre T/L, see Table 1 for a description of
(ODD). Herpertz et al. (2001), instead, found reduced NS-SCRs in these measures). Other studies, instead, argued that hypo-arousal at
children with ADHD and conduct disorder (CD), compared to controls, rest, in ADHD, may be due to reduced activation of the SNS. For ex-
but no significant difference between ADHD and controls, and they did ample, Crowell et al. (2006) found that pre-schoolers with ADHD
not replicate these results on a larger sample (Herpertz et al., 2003) showed increased baseline pre-ejection period (PEP) length, and the
Among the studies reporting hyper-arousal, the method of mea- same interpretation was proposed by Beauchaine et al. (2001), who
suring skin conductance was not described in detail by Mukhopadhyay found signs of reduced SNS activity in boys with ADHD and co-morbid
et al. (1997), while Hermens et al. (2005a) found signs of hyper-arousal CD, (i.e., increased PEP length and reduced high-frequency respiratory
in female adolescents with ADHD, using the difference of SCL from the sinus arrhythmia power; HF-RSA). However, the absence of a sig-
beginning of the resting period (instead of mean SCL) as a measure of nificant difference between ADHD without CD and controls, indicates

185
 Table 1
Description of measures which were used in the studies included in the review, including their relation with functioning of the autonomic nervous system and the methodology usually used to collect and extract these
measures.
A. Bellato, et al.

Domain Measure Acronym Significance & methodology Parameters ANS indicator Number of reviewed
studies using the
measure

EDA Skin Conductance Level SCL An electrical potential is applied to two electrodes, placed next to Mean SCL, Higher SCL: increased sympathetic arousal 14 studies (resting)
each other on the hand palms. The electrical current, flowing Change (slope) of SCL 7 studies (task)
between the electrodes, is measured as an index of skin over time
conductance.
SCL is a tonic measure of skin conductance and detects slow
changes over time.
Non-specific Skin Conductance ns-SCR Methodology is the same as for SCL. Ns-SCRs are spontaneous Number of ns-SCR/Fs Higher ns-SCR/Fs: increased phasic arousal 7 studies (resting)
Response/Fluctuation ns-SCF changes (difference in SC from a baseline level) in skin Mean ns-SCR/Fs or responsivity, but not specific to task- 1 study (task)
conductance, which are not linked to any task-related event or related stimuli or events
stimulus. They can be measured during baseline, resting or task.
Skin Conductance Response SCR Methodology is the same as for SCL and ns-SCRs. Similarly to ns- Mean SCRs Higher SCRs: increased phasic arousal or 14 studies (task)
SCRs, SCRs are phasic changes (measured as a difference in SC responsivity to specific events or stimuli
compared to a baseline) but they are measured in response to a
specific event or stimulus
Heart rate Heart rate HR Measured on electrocardiogram (ECG), it is calculated as the Mean HR or IBI Higher HR: hyper-arousal 7 studies (resting)
average number of beats per minute (BPM). Increased stimulus-locked IBI: deceleration 19 studies (task)
Inter-beat interval (IBI) around a specific task stimulus represents Reduced stimulus-locked IBI: acceleration
a phasic change in heart rate (acceleration or deceleration)
Standard deviation of normal-to- SDNN or R-R- Measured by calculating the average change in the duration of the SDNN Higher SDNN: increased HRV: increased PNS 2 studies (resting)
normal intervals (Heart Rate interval-SD interval (in msec) between consecutive heart-beats (R-R intervals) functioning: hypo-arousal 4 studies (task)
Variability; HRV) and deriving the standard deviation of the N-N (or R-R) intervals
over a time section of the ECG.

186
Root Mean Square of Successive RMSSD A measure of HRV which is measured as the average root square RMSSD Higher RMSSD: increased HRV: higher PNS 5 studies (resting)
Differences (HRV) of the interval between successive peaks in ECG. Compared to functioning: hypo-arousal 3 studies (task)
SDNN, it is considered more reliable in measuring HRV.
Normal-to-normal intervals NN50 (pNN50) Measured on ECG as the number of pairs of successive IBI Number of NN50 Higher number of NN50 (or higher pNN50): 2 studies (resting)
> 50 ms (% of) intervals which are different by 50 msec or more. pNN50 is the increased HRV: higher PNS functioning:
proportion of NN50, in relation to the entire number of RR hypo-arousal
intervals.
Pre-ejection period PEP Measured as the interval between the electrical depolarization of PEP length Increased PEP length: reduced SNS 7 studies (resting)
the left ventricle and the beginning of blood pumping from the functioning: hypo-arousal 8 studies (task)
same location. It reflects the time needed for the left ventricle to
contract and pump blood, after the cardiac valves have closed. It
is mainly influenced by activation of the SNS (it reduces when
SNS activates).
Respiratory sinus arrhythmia RSA Represents the association between cardiac functioning and RSA Increased RSA: increased PNS functioning: 10 studies (resting)
respiratory mechanisms. It is measured as the variability of heart hypo-arousal 10 studies (task)
rate associated with breathing, with heart rate accelerating during
inspiration and decelerating during expiration. Although it is
likely to affect PEP, RSA is more related to PNS, being an index of
cardiac vagal tone.
Low frequency power LF As a frequency domain measure, it represents the amount of LF power Increased LF power: increased baroreflex 5 studies (resting)
spectral power between 0.04 - 0.15 Hz on the Fast Fourier effect: increased HRV 4 studies (task)
Transform (FFT) spectrum of HRV. It is indicative of the
baroreflex which is a modulation (acceleration or deceleration) of
HR in situations when blood pressure (BP) is too low or high
(respectively), with the aim of changing BP levels through
modulating HR.
High frequency power HF Similarly to LF, it is a frequency domain measure of HRV which HF power Increased HF power: increased PNS 5 studies (resting)
analyses activity in the 0.15–0.40 Hz range. It has been linked to functioning: hypo-arousal 2 studies (task)
(continued on next page)
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Number of reviewed that these findings may be specifically related to the co-occurring pre-
sence of ADHD and CD. In support of this idea, Herpertz et al. (2003)

5 studies (resting)

1 studies (resting)
studies using the
found signs of hypo-arousal, measured through mean heart rate, in both

2 studies (task)

2 studies (task)
CD and ADHD + CD, but not in ADHD-only, and also Griffiths et al.
(2017) found signs of imbalance between SNS and PNS functioning at
measure

rest (i.e., increased low-frequency/high-frequency ratio; LF/HF) in

ADHD (particularly in males), but this was mainly predicted by oppo-
sitional problems and not ADHD per se. This suggests that HR measures

allocation of mental resources to the external

Phasic pupil size dilations: increased arousal,

literature, regarding the association between
of ANS functioning may be specifically affected by the presence of co-

spontaneous or in response to a stimulus

(however a strong debate is going on in
Reduced LF/HF ratio: parasympathetic

morbid symptoms, such as conduct and oppositional behaviours, rather

Increased tonic pupil size: increased

Increased LF/HF ratio: sympathetic

than inattention or hyperactivity. However, since some studies (e.g.,

Herpertz et al., 2001) did not find any difference between ADHD,
ADHD + CD and controls, further research is needed to establish how
the presence of comorbid CD/ODD affect HR measures in ADHD, at rest.
LF/HF ratio and ANS)

Among the studies reporting hyper-arousal, Rukmani et al. (2016)

found evidence of atypical SNS-PNS balance in ADHD, by analysing
ANS indicator

environment

measures of HRV, but only 10 children were included in each of the

dominance

dominance

clinical and control groups, undermining the generalisability of the

findings. Signs of hyperarousal were found by Shibagaki and Furuya
(1997), who found that more children and adolescents with ADHD,
than controls, showed reduced RSA, and concluded that hyper-arousal
Stimulus-or event-locked
Mean pupil size (tonic)

could be caused by reduced vagal tone. Leikauf et al. (2017) compared

changes in pupil size

cognitively defined biotypes of ADHD described as ‘impulsive-cogni-

tion’ and ‘inattentive-cognition’ on HR and EDA measures. Overall, they
LF/HR ratio
Parameters

found increased mean HR in in the ‘impulsive-cognition’ group, com-

(phasic)

pared to controls, while there was no difference between the ‘in-

attentive-cognition’ group and typically developing controls. This result
suggests that specific traits of ADHD, e.g., impulsivity, may be asso-
Frequency range (see above for specific ranges in Hz). It has been

situated at a close distance (due to narrowing of visual attentional

used as an index of the balance between SNS and PNS functioning
cardiac-vagal activity, representing parasympathetic modulation

ciated with different profiles of autonomic functioning at rest.

effects, including changes in environmental luminance or body
movements. Pupil constrictions are generally associated with
optometrist tools, carefully considering for any confounding

Many studies reporting no group differences between ADHD and

increase of brightness and processing of visual information
It is the ratio between spectral power in the low and high

Pupil size or diameter is measured using eye-tracking or

typical controls measured mean HR, suggesting that calculating mean

HR over a certain period of time may not be sensitive enough to detect
signs of arousal dysregulation in ADHD. Although none of the studies
on adults found group differences on HR measures (Lackschewitz et al.,
2008; Oliver et al., 2012), further research is needed, given the paucity
of research assessing adults.

3.1.3. Pupillometry
Significance & methodology

A clear picture of ANS functioning at rest, derived through pu-

(sympathovagal balance).

pillometry, did not emerge from this review, since the only study that
focused on this ANS measure (Kara et al., 2013) did not find any group
differences between controls and ADHD, or between ADHD and
ADHD + ODD (Table 4).
of arousal.

focus).

3.1.4. Summary of resting state findings

Of 31 studies investigating autonomic arousal at rest, 12 showed
signs of hypo-arousal, four of hyper-arousal and a large number of
studies (15) did not report any significant group differences. Further
Acronym

research using all these measures of autonomic arousal in parallel is

LF/HF

needed to understand the heterogeneous findings. Methodological

choices, e.g., in the design of the resting-state period as a proper period
of resting or as a short break between cognitive tasks, and the nature of
the sample, e.g., including children with comorbid symptoms, may also
Low/high frequency power

be at the basis of non-significant or heterogeneous findings, which will

be further discussed in paragraph 4.5.

3.2. Cognitive, reinforcement or socio-emotional tasks

Pupil diameter
Measure

Temporary changes in autonomic measures of ANS functioning in

response to a particular stimulus reflect phasic modulation of the ANS
Table 1 (continued)

and may be measured to identify the presence of atypicalities in ANS

PUPILLOMETRY

regulation. Phasic stimulus-locked autonomic responses, reviewed in

this section, were measured using EDA, HR and pupillometry and
during paradigms with different designs, such as cognitive tasks (e.g.,
Domain

attention, inhibitory control, working memory, alternative forced-

choice, and sustained attention), paradigms manipulating rewards or

187
 A. Bellato, et al.

Table 2
Summary of results from studies measuring electrodermal activity during resting-state.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect Hypo/
hyper/none

Barry et al., 2009 30 30 Children and Resting state 5 min. Mean SCL ADHD < CTRl Hypo (males only)
adolescents
Barry et al., 2012c 18 18 Children and Resting state 3 mins Mean SCL ADHD < CTRL Hypo
adolescents
b,d
Beauchaine et al., 2001 17 22 typical; Adolescents Baseline 5 min., ns-SCRs ADHD < CTRL Hypo (males only)
20 CD before task start
Clarke et al., 2013 104 67 Children and Resting state Mean SCL ADHD, defined as either excess EEG theta power OR Hypo (males only)
adolescents 10.5 min. excess beta EEG power, < CTRL
Crowell et al., 2006b 18 20 Pre-school children Resting state 5 min. Ns-SCRs ADHD and ODD < CTRL Hypo
Dupuy et al., 2014 40 40 Children and Resting state 5 min. Mean SCL ADHD < CTRL Hypo (females only)
adolescents
Hermens et al., 2004 35 35 Adults Resting state 2 min. Mean SCL ADHD < CTRL Hypo (females only)
Hermens et al., 2005a 70 70 Adolescents Resting state 2 min. Slope of SCL (rate of ADHD < CTRls on rate of change in SCL Hyper (females only)
change), ns-SCRs
Hermens et al., 2005bc 34 34 Adolescents Resting state 3 mins Slope of SCL (rate of No group differences when ADHD off-medication None
change), ns-SCRs

188
Herpertz et al., 2001a 47 (26 ADHD + CD) 21 Children and Resting state 3 mins Mean SCL, ns-SCRs No group differences on mean SCL. Hypo (males only,
adolescents ADHD + CD < CTRl on ns-SCRs ADHD + CD only)
a
Herpertz et al., 2003 28 ADHD, 20 CD, 50 29 Children and Resting state 3 mins Mean SCL, ns-SCRs No significant differences between groups on SCL or ns- None (males only)
ADHD + CD adolescents SCRs
b
Iaboni et al., 1997 18 18 Children and Resting state 5 min. Mean SCL No difference between groups on mean SCL None (males only)
adolescents
Lazzaro et al., 1999 54 54 Children and Resting state 2 min. Mean SCL, ns-SCRs ADHD < CTRL Hypo
adolescents
McQuade and Breaux, 23 ADHD, 7 subthreshold 31 Children and Resting 3 mins Mean SCL No significant differences None
2017a,d ADHD adolescents
McQuade et al., 2017a,d 23 ADHD, 7 subthreshold 31 Children and Resting 3 mins Mean SCL No significant differences None
ADHD adolescents
Mukhopadhyay et al., 1997 10 10 Children and Unclear SCL fluctuations ADHD > CTRL Hyper
adolescents
COUNT (Hypo:hyper:none) 9:2:5

a
Resting state measured before, during or after a cognitive task (see Tables 5–7 for cognitive task findings).
b
Resting state measured before, during or after a reward/reinforcement task (see Tables 8 and 9 for reward/reinforcement task findings).
c
Measured effects of an intervention (see Table 11 for results of intervention).
d
Resting state measured before, during or after a socio-emotional task (see Table 10 for socio-emotional task findings).
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
 Table 3
Summary of results from studies measuring heart rate during resting-state.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect Hypo/
A. Bellato, et al.

hyper/none

Beauchaine et al., 17 ADHD + 20 22 Adolescents Resting state 5 minutes PEP, RSA ADHD + CD > ADHD & CTRL on PEP length. Hypo (males only, some
2001b ADHD + CD pre-task and 2.5 minutes ADHD + CD < CTRl on HF RSA power effects only for
between task blocks/tasks ADHD + CD)
Crowell et al., 2006a 18 ADHD + ODD 20 Pre-schoolers Resting 5 min. PEP, RSA ADHD + ODD > TD on baseline PEP length Hypo
ADHD + ODD = TD on baseline RSA
De Carvalho et al., 28 28 Children Resting state 20 mins Frequency (LF, HF, LF/ ADHD = TD on HF, LF and LF/HF ratio. ADHD > TD on NN50 Hypo
2014 HF), RMSSD, NN50, ADHD > TD on Poincarre T/L
pNN50
Griffiths et al., 229 244 Children and Resting state 2 min. RMSSD, HF, LF, LF/HF ADHD = CTRl on heart rate. ADHD > CTRl on LF/HF ratio Hypo (males only)
2017a adolescents Correlations with comorbid symptoms
Herpertz et al., 21 ADHD, 26 21 Children and Resting state 3 mins Mean heart rate ADHD + CD = ADHD = TD on resting mean HR None (males only)
2001a ADHD + CD adolescents
Herpertz et al., 28 ADHD, 20 CD, 50 29 Children and Resting state 3 mins Mean heart rate ADHD + CD, CD < CTRl mean heart rate. No difference Hypo (males only,
2003a ADHD + CD adolescents between ADHD and CTRL effects in ADHD + CD
and CD only)
Iaboni et al., 1997b 18 18 Children and Resting state 5 min. Mean heart rate No group differences None (males only)
adolescents
Jennings et al., 40 26 Children and Resting state (2 min.) Mean heart rate (inter- Non-significantly faster heart rate in ADHD than CTRL None (males only)
1997a adolescents beat interval)
Lackschewitz et al., 18 18 Adults Resting state 15 minutes SDNN, RMSSD, Power No group differences None
2008d Spectral Density, LF, HF,
LF/HF
Leikauf et al., 112 56 Children and Resting state 2 min. Heart rate, RMSSD Two ADHD biotypes defined by cognitive profile. ‘Impulsive Hyper (specific to one
2017a,c adolescents cognition’ biotype showed faster mean HR. ‘Inattentive bio-type)

189
cognition’ biotype showed no difference in HR. No effects for
RMSSD
McQuade and 23 ADHD, 7 subthreshold 31 Children and Resting 3 mins RSA, RDA reactivity No group differences None
Breaux, 2017 a,d ADHD adolescents (RSA-R: difference
between baseline RDA
and task RSA)
Musser et al., 2011d 32 32 Children Resting state 2 min. PEP, RSA No group differences None
Musser et al., 2013d 21 ADHD low social 75 Children and Resting state (length not PEP, RSA No group difference None
function; 54 ADHD high- adolescents reported)
social function
Musser et al., 42 44 Children and Resting state (length not PEP, RSA No group difference None
2018d,e adolescents reported)
Oliver et al., 2012 N/A 22 low-ADHD- Adults Driving simulation task Heart period, RSA, PEP No group differences None
traits; 20 high-
ADHD traits
Rukmani et al., 10 10 Children and Resting state 15 min. Heart rate, Mean & SD of ADHD < CTRl on RMSSD, pNN50, SDNN, HF. ADHD > CTRl Hyper
2016 adolescents NN intervals, RMSSD, on LF/HF ratio
NN50 (n, %), LF, HF, LF/
HF
Shibagaki and 18 49 Children and Resting state 3 mins RSA More ADHD children than controls showed decreasing or no None
Furuya, 1997 adolescents rhythmicity, derived from RSA.
No group differences on RSA.
Tenenbaum et al., 69 48 Children and Resting state 2 min. PEP, Mean RSA No group differences None
2018b adolescents
Wang et al., 2013 Cohort of 88 (ADHD traits Pre-schoolers Baseline Power spectrum analysis Male pre-schoolers with more inattentive/hyperactive traits Hypo
measured) of HRV showed lower sympathetic and higher parasympathetic activity (high non-clinical traits,
males only)
(continued on next page)
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

ANS group effect Hypo/ reinforcement (e.g., reward versus penalty, reinforcement versus no or
neutral feedback) and socio-emotional paradigms (social cognition and
emotion recognition or processing). We begin by reviewing studies that
measured ANS during cognitive tasks.
hyper/none

3.2.1. Cognitive tasks

6:2:12
None

3.2.1.1. EDA. Of 12 studies recording EDA during cognitive tasks,

seven reported evidence of ANS hypo-activation in ADHD, one of
hyper-activation and four of no group differences (Table 5).
Of the studies reporting hypo-activation of the ANS, three used ei-
ther passive and/or active auditory attention tasks (Herpertz et al.,
2001, 2003; Shibagaki et al., 1993). Shibagaki and Furuya (1997)
further demonstrated that, in the passive task, the difference between
ADHD and controls was greatest at the beginning of the tasks, in-
dicating difficulties in adapting to a novel experimental situation. The
studies of Herpertz et al. (2001; 2003), described in paragraphs 3.1.1
and 3.1.2, similarly found reduced skin conductance responses (SCRs)
and delayed habituation of SCRs to auditory tones during a passive
attention task, but this effect was only significant when comparing
COUNT Hypo:hyper:none

children with ADHD + CD to typical controls (no differences were

No group differences

found between controls and children with ADHD-only). Since the other
studies did not measure the influence of CD, it is not clear whether
Main findings

those effects were mainly driven by co-occurring CD as well.

Resting state measured before, during or after a reward/reinforcement task (see Tables 8 and 9 for reward/reinforcement task findings).

Two other studies reported hypo-activation of the ANS during sus-

tained attention tasks (Lawrence et al., 2005; O’Connell et al., 2004).
Lawrence et al. (2005) found reduced mean SCL in children and ado-
lescents with ADHD during an adapted version of the Continuous Per-
formance Task (CPT), suggesting reduced activation of the ANS, while
O’Connell et al. (2004) found atypical phasic modulation of skin con-
Resting state measured before, during or after a socio-emotional task (see Table 10 for socio-emotional task findings).

ductance in ADHD children and adolescents during a different form of

ANS measure

CPT, the Sustained Attention to Response Test (SART). Specifically,

while in control children SCRs following commission errors were in-
Resting state measured before, during or after a cognitive task (see Tables 5–7 for cognitive task findings).
RSA

creased relative to correctly withheld trials, this difference was not

significant in ADHD, suggesting a lack of ANS responsivity to errors in
ADHD. Considering firstly, the role of the ACC in error processing
(Brown and Alexander, 2017; Sellaro et al., 2015), secondly, evidence
An erratum for this paper (Musser et al., 2018) reported the same results found in the original.
Resting state 2 min.

of atypical function of brain systems that support error processing in

ADHD (Johnstone et al., 2013) and thirdly, the links between ACC and
ANS (Critchley and Garfinkel, 2018), the evidence reviewed here sug-
Paradigm

gests that ANS dysfunction may contribute to abnormal error proces-

sing in ADHD.
Measured effects of an intervention (see Table 11 for results of intervention).

One other study reported evidence of reduced EDA in ADHD

(Johnstone et al., 2010) using a hybrid flanker go/no-go task designed
Children and

to challenge response conflict processing and inhibitory control. In

adolescents
Age group

addition, these authors manipulated the effort required to pay attention

to the stimulus stream by degrading the central target stimulus by ei-
ther 0 %, 30 % or 60 %. Children and adolescents with ADHD showed
significantly reduced mean SCL compared with matched typical con-
trols in the non-degraded condition of the task but equivalent SCL in the
Control n

60 % degraded condition, suggesting hypo-activation of the ANS in the

easier condition but typical arousal levels in the most difficult condi-
127

tion. James et al. (2016) found significantly reduced SCL in the baseline
condition of a 4-choice RT task in the ADHD group compared with
controls. In a faster paced condition with incentives for correct re-
sponses, no significant group differences were reported and the ADHD
group showed a steeper increase in SCL from the baseline to the fast-
incentive condition than the control group. A further analysis of this
ADHD n

effect can be found in the next paragraph, where the impact of reward
on cognitive performance is discussed. Together, these findings suggest
116

that ADHD children may perform better in conditions with a faster

Table 3 (continued)

event rate or that are perceptually more demanding, compared with

Ward et al., 2015a
First author, year

slower, more monotonous tasks with very few cognitive or perceptual

challenges.
Contrary to these findings, one study (Satterfield et al., 1984) re-
ported increased SCL in children with a diagnosis of Hyperactivity
d
b
a

(DSM-III diagnosis of Attention Deficit with Hyperactivity; ADD/H)

c

190
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Table 4
Summary of results from studies measuring pupil size measures during resting-state.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect Hypo/hyper/none

Kara et al., 2013 32 24 Children Resting-state Pupil diameter ADHD = CTRL None
COUNT (hypo:hyper:none) 0:0:1

passively attending to clicks delivered at either a slow or fast rate bi- and adolescents with ADHD showed a profile of greater RT and RTV
naurally, which would suggest increased ANS activation, while other (but not errors) compared with controls, as well as greater power in
studies reported no group differences in EDA recorded during a cog- the.10 Hz heart rate frequency and poorer on-task behaviour (1999),
nitive task. Mayer et al. (2016), for example, used the Contingent Ne- suggesting less effort allocation and weaker ANS regulation. Using a
gative Variation (CNV) to assess response preparation during an audi- similar paradigm, Griffiths et al. (2017) found greater LF/HF ratio
tory go/no-go task in adults with ADHD. Although increased reaction- during rest and task in ADHD children that, similarly to Borger et al.
times variability (RTV) and reduced CNV amplitude were found in (1999), predicted worse task performance. The studies presented so far
ADHD, the groups did not differ on mean SCL or SCRs to the cue-go seem to converge on a profile of hypo-activation of the ANS in ADHD
stimulus interval. Moreover, Dykman et al. (1982) found no group during cognitive tasks that require response control and sustained at-
differences in mean SCL in the inter-trial interval of a stimulus-response tention. However, one study (Shibagaki and Furuya, 1997) used passive
mapping task, in which children with a diagnosis of ADHD-Hyperactive and active auditory listening tasks and reported no group differences in
Subtype (DSM-III) with or without Reading Disorder, were required to HR during the tasks, although the ADHD group showed reduced RSA
learn a stimulus-response (S-R) mapping to a criterion level of perfor- variability during a resting state condition. The analytic approach used
mance, at which point the S-R mapping switched. Further studies in this paper relied upon calculating the frequency of RSA sub-types
(McQuade et al., 2017; McQuade and Breaux, 2017) did not find any within the ADHD and control samples; the methodological approach is
association between ADHD symptoms and SCL collected during a bat- therefore not comparable to other studies in this review that have
tery of executive function, cognitive and socio-emotional tasks in measured HR in relation to attention.
children and adolescents with different levels of ADHD traits, but found The only study investigating error monitoring found signs of hypo-
a significant positive association between SCL and internalising symp- activation of the ANS in ADHD during a task that required participants
toms. to report the global or local shape in a global/local array, interpreted as
In summary, the majority of studies measuring EDA during a cog- reduced sensitivity to error processing (Groen et al., 2009). Also, all
nitive task have found significantly reduced EDA (either mean SCL or groups (controls, ADHD-unmedicated and ADHD-medicated) were
target-locked SCR) in children and adolescents with ADHD, reflecting more accurate and gave slower responses in the feedback blocks com-
hypo-activation of the ANS, although four studies reported no group pared with no feedback, and show similar post-error slowing, while
difference and one reported evidence of hyper-activation. The studies stimulus-locked HR decelerations were greater on error trials than
differ in task design and sample composition; factors that will be dis- correct trials in controls and medicated-ADHD, but not in unmedicated
cussed further in paragraph 4.5. ADHD. Broadly, these findings indicate reduced autonomic reactions to
errors in ADHD children when unmedicated, supporting evidence pre-
sented in paragraph 3.2.1.1 and suggesting that the links between ACC
3.2.1.2. Heart rate. Twelve studies analysed autonomic functioning
and ANS may be atypical in ADHD, undermining error processing.
during cognitive tasks in ADHD by collecting cardiac measures
In another study (Leikauf et al., 2017), the authors used a battery of
(Table 6). Among these, seven found signs of hypo-activation of the
cognitive tasks, including Go/no-go, CPT, attention switching task,
ANS and five found no group differences between ADHD and typical
maze task, verbal memory recall task, verbal inference task, motor
controls (none of the included studies reported hyper-activation).
tapping task, digit span and choice RT task, and identified two cognitive
In a test of the state regulation hypothesis which suggests that
subtypes of ADHD, namely “impulsive” and “inattentive”, based on
ADHD children are less able to increase effort allocation during cog-
their performance to the tasks. A significant difference on HR measures
nitively boring or sustained tasks, Borger and van der Meere (2000)
was found only for the impulsive-cognition subtype at rest (see para-
measured changes in heart rate (inter-beat interval; IBI) prior to and
graph 3.1.2). In contrast, the inattentive-subtype had longer/more
following the presentation of go and no-go stimuli. The authors found
variable RTs and omission errors and lower EEG beta power, but HR did
significantly reduced pre-stimulus HR decelerations in the ADHD group
not differ from controls. These results suggest the importance of con-
in the slow but not in the fast event rate condition, and a delay in the
sidering subtypes within the ADHD diagnostic category, which may
onset of HR acceleration in the slow condition, suggesting a lack of ANS
better explain the heterogeneity of results about arousal regulation
regulation related to reduced motor preparation in the slow condition.
deficits in ADHD.
In addition, power in the.10 Hz IBI frequency range was greater in
Five studies (Dykman et al., 1982; Keage et al., 2006; McQuade and
ADHD than controls, in the slow condition, indicating less effort allo-
Breaux, 2017; Perrin et al., 2014; Ward et al., 2015) used one or more
cation during this less challenging task condition. There were no group
tasks assessing executive functions. Two studies reported some evi-
differences in HR on no-go trials. Similarly, Jennings et al. (1997)
dence of ANS dysfunction in ADHD. Dykman et al. (1982) measured HR
measured HR IBIs with respect to a go-stimulus in a stop signal task and
in hyperactive children (with or without reading disabilities) and ty-
found slower go RT and increased standard deviation of RTs (RTs-SD) in
pically developing controls during a rewarded stimulus-response
the ADHD group. Furthermore, although control children showed
learning task. The hyperactive children showed reduced HR accelera-
longer IBIs (i.e., heart rate deceleration) preceding successful than
tion on trials that were less likely to lead to reward which the authors
failed inhibitions, suggesting greater autonomic control on trials when
interpreted as reduced effort allocation when rewards are improbable.
the response was successfully withheld, ADHD children did not show
Ward et al. (2015) measured short term memory and found that chil-
this effect. Overall, both studies assessing response inhibition found
dren with higher ADHD symptoms showed worse task performance and
signs of hypo-arousal in ADHD, but the atypical features in ADHD were
reduced RSA withdrawal from baseline/rest to the task, which is a sign
more closely related to response preparation and regulation, rather
of reduced functioning of the PNS when switching from a less- to a
than response inhibition.
more-attentional demanding situations. However, a positive association
Signs of ANS hypo-activation have been reported in ADHD by stu-
between good task performance and ADHD symptom scores was also
dies investigating sustained attention using variants of a CPT. Children

191
 A. Bellato, et al.

Table 5
Summary of results from studies measuring electrodermal activity during cognitive tasks.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect
Hypo/hyper/none

Dykman et al., 1982 10 Hyperactive, 10 10 Children and S-R contingency task Mean SCL 30 seconds pre and No differences between groups None
Reading Disabled, 10 adolescents post-test and during intertrial
Hyp + RD interval
Herpertz et al., 21 ADHD, 26 ADHD + CD 21 Children and Auditory attention task SCR to stimuli; No differences between ADHD and typically Hypo (only in
2001a adolescents (passive) and startle stimulus Habituation (reduction in SCR) developing controls. Differences were specific to ADHD + CD)
task to stimuli ADHD + CD
Herpertz et al., 28 ADHD, 20 CD, 50 19 Children and Auditory attention task SCR to stimuli; No differences between ADHD and typically Hypo (only in
2003a ADHD + CD adolescents (passive) and startle stimulus Habituation (reduction in SCR) developing controls. Differences were specific to ADHD + CD and CD)
task to stimuli ADHD + CD and CD
James et al., 2016 73 + 75 siblings 72 + 72 Adolescents and 4-choice RT Fast Task SCL and SCRs ADHD < CTRl on SCL during the slow-event rate Hypo
siblings young adults block with no incentives, and not during fast-event
rate with incentive).
ADHD = TD on SCRs
Johnstone et al., 20 20 Adolescents Flanker task with 3 levels of Mean SCL ADHD < TD on SCL (only in the non-degraded Hypo
2010 degradation of the central condition)
stimulus
Lawrence et al., 18 18 Children and CPT-AX Mean SCL, stimulus-locked ADHD < controls on SCL during first session Hypo

192
2005b adolescents SCR (difference reduced with medication), > controls
during second session.
ADHD < controls on SCR (only in test phase 2)
Mayer et al., 2016 23 22 Adults Auditory cued go/no-go (eyes SCL No differences between groups None
closed)
McQuade and 23 ADHD, 7 subthreshold 31 Children and CANTAB, impossible puzzles Mean SCL SCL during the tasks was associated with internalising None
Breaux, 2017a ADHD adolescents symptoms, but not ADHD symptoms.
McQuade et al., 23 ADHD, 7 subthreshold 31 Children and CANTAB, impossible puzzles Mean SCL SCL during the tasks was associated with internalising None
2017a ADHD adolescents symptoms, but not ADHD symptoms.
O’Connell et al., 15 15 Children and Sustained Attention to SCR and SCL No difference in SCR to commission errors and Hypo
2004 adolescents Response Task withholds in ADHD (difference was present in
controls).
ADHD < controls on SCR to commission errors
Satterfield et al., 138 60 Children and Passive attention: clicks Mean SCL Hyperactive > TD on SCL Hyper
1984 adolescents delivered binaurally while
watching a video.
Shibagaki et al., 18 49 Adolescents Passive and active listening Mean SCR ADHD < controls on SCR, especially during the first Hypo
1993 task trials of the passive listening task, and on the entire
active listening tasks
COUNT (Hypo:hyper:none) 7:1:4

a
Resting-state measures obtained before, during or after the cognitive task (see Tables 2–4 for resting state findings).
b
Measured effects of an intervention (see Table 11 for results of intervention).
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
 Table 6
A. Bellato, et al.

Summary of results from studies measuring heart rate during cognitive tasks.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group
effect Hypo/
hyper/none

Borger et al., 1999 21 16 Children and Sustained attention task Mean HR, No differences between groups on mean HR. Hypo (males
adolescents Mean power in the.010 Hz ADHD > CTRl on 0.10 Hz component (ADHD: reduced only)
component effort allocation)
Borger and van der 27 22 Children and Go/no-go task (response inhibition) Mean IBI around go and no-go No difference between groups on mean IBI. Hypo (males
Meere, 2000 adolescents stimuli; ADHD < CTRl on HR deceleration (in the slow only)
Mean power in the.10 Hz condition, not in the fast).
component ADHD > CTRl on delay in IBI acceleration after go
stimulus presentation, in the slow condition.
ADHD = CTRl on HR deceleration after No-Go stimuli,
when compared to Go.
ADHD > CTRl on.10 Hz component (slow condition),
indicating less effort allocation
Dykman et al., 1982 10 Hyperactive, 10 10 Children and S-R task (executive functions, set HR rate, inter-trial interval No differences between groups on HR. Hypo
Reading Disabled, 10 adolescents shifting) changes in HR Hyperactive and RD > CTRl on inter-trial interval
Hyp + RD changes in HR (reduced effort allocation in the clinical
groups)
Griffiths et al., 2017a 229 224 Children and CPT-IP (sustained attention) RMSSD, HF, LF, LF/HF ADHD > TD on LF/HF during the task (lower LF/HF Hypo
adolescents during task predicted better performance). No differences
between groups in RMSSD, HF or LF
Groen et al., 2009 16 + 16 18 Children and Global/local task (error processing) Stimulus-locked IBI Unlike TD and ADHD-with-MPH, ADHD-unmedicated did Hypo

193
adolescents not show increased HR deceleration/acceleration effect
on error trials, compared to correct trials
Jennings et al., 40 26 Children and Stop Signal Task (response inhibition) Go-stimulus-locked IBI Unlike CTRL, ADHD did not show longer IBIs preceding Hypo (males
1997a adolescents successful inhibition, than failed inhibitions (heart rate only)
deceleration before successful inhibition)
Keage et al., 2006 129 129 Children and Trail making, verbal memory Mean HR No significant differences between groups None
adolescents (executive functions; working
memory)
Leikauf et al., 112 56 Children and Go/no-go, switching, CPT, maze, Heart rate, RMSSD No significant differences between groups None
2017a,b adolescents verbal memory recall, verbal
inference, motor tapping, digit span,
choice RT
McQuade and 23 ADHD, 7 subthreshold 31 Children and CANTAB (executive functions), RSA, RDA reactivity (RSA-R: No group differences on HR measures. None
Breaux, 2017a ADHD adolescents impossible puzzles difference between baseline
RDA and task RSA)
Perrin et al., 2014 19 children; 34 children; Children and Tower of London (executive functions) Amplitude and latency of HR No significant differences between groups None
20 adults 22 adults adults acceleration and deceleration
Shibagaki and 18 49 Children and Passive and active auditory attention HR, stimulus-locked HR No significant differences between groups None
Furuya, 1997a adolescents task acceleration and deceleration
Ward et al., 2015a 116 127 Children Forward and backward Short Term RSA No significant differences between groups. Hypo
Memory Task (executive functions; Higher probability of ADHD diagnosis and poor STM
working memory) performance was associated with lower RSA withdrawal
from rest to task.
COUNT (Hypo:hyper:none) 7:0:5

a
Resting-state measures obtained before, during or after the cognitive task (see Tables 2–4 for resting state findings).
b
Measured effects of an intervention (see Table 11 for results of intervention).
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

found, but only when RSA withdrawal from rest to task was increased. of impairments found in ADHD.
This finding suggests that reduced responsivity of the PNS branch of the
autonomic system in the transition from a baseline rest period to a task 3.2.2. Reward/reinforcement
may undermine task performance in ADHD. Eleven studies included in this review, measured the association
The remaining three studies measuring executive functions (EF) between reward/reinforcement and ANS functioning, in ADHD, by in-
found no group differences in cardiac measures of ANS activity. vestigating the effects of rewards/reinforcers on some aspect of cogni-
McQuade and Breaux (2017) used several experimental paradigms, tion, such as attention or inhibitory control, or the effect of changing
including an executive function battery and other socio-emotional tasks the frequency or magnitude of rewards and penalties.
(see below, paragraph 3.2.3) but only found effects relating to EDA (see
paragraph 3.2.1.1), not HR. In a study designed to measure HR and 3.2.2.1. EDA. Two studies (out of eight using EDA) found signs of
electrophysiological responses (P3a amplitude and latency) to dis- hypo-activation of the ANS in ADHD, one found signs of hyper-
tractor stimuli in a working memory task, Keage et al. (2006), found no activation and five reported no group differences (Table 8).
HR differences between groups. Perrin et al. (2014) measured HR in a Iaboni et al. (1997) were among the first to test the hypothesis that
wide age range of children and adults with ADHD and controls while the Behavioural Inhibition System (BIS), introduced by Gray (1981), is
they performed an age-adjusted version of the Tower of London (TOL) deficient in ADHD. In their study, children performed a simple RT task
task but found no significant differences between groups on HR accel- and were initially rewarded before the reward contingencies were ex-
eration or deceleration, or task performance. The authors conclude that tinguished and then reintroduced. SCL (which was averaged for each
ADHD may not suffer from an EF planning deficit and that the task may block as an index of SNS activity and a measure of BIS activation) in-
have been excessively engaging, since feedback was given on each trial, creased as soon as rewards were extinguished in controls, but not in
so this may have reduced any arousal regulation difficulties in ADHD. children with ADHD, indicating inefficient regulation of the ANS in
Overall, the findings suggest that ANS function is generally typical in response to a change in reinforcement contingencies (see results from
ADHD during EF tasks, although this may depend on task complexity. HR measures in paragraph 3.2.2.2). A similar version of this task was
used by Beauchaine et al. (2001) who reported no group differences
3.2.1.3. Pupillometry. Only two studies measured pupil size during between children with ADHD, ADHD + CD and controls on SCRs to
cognitive tasks (Table 7). One reported signs of hypo-arousal in reward extinction. The reason for the discrepant findings between these
ADHD (Wainstein et al., 2017), while the other reported no studies is unclear.
differences between ADHD and controls (Karatekin et al., 2010). Besides reporting reduced SCL during a no-incentive slow condition
Pupil size was reduced among off-medication children and adoles- of a forced-choice response task, James et al. (2016) (see paragraph
cents with ADHD performing a visuo-spatial working memory task 3.2.1.1), found a significant increase in SCL and an improvement in task
(Wainstein et al., 2017), compared to controls, but the difference was performance in a fast, incentivised condition in the ADHD group only,
absent when on-medication (see paragraphs 3.3 and 4.6 for a discussion not in control children. The effect of this was to equate the SCL and
about the effects of medication). Moreover, a correlation between performance of the ADHD and control group suggesting that ANS ac-
within-trial pupil size and performance (i.e., accuracy and RTV) was tivity and attention are enhanced by faster event rates and incentives,
found in the ADHD group, suggesting the presence of signs of hypo- although these two factors cannot be dissociated within that particular
activation of the ANS in ADHD and specific difficulties in allocating a paradigm. Conversely, two studies measuring relationships between
constant and appropriate level of attentional resources during tasks inhibitory control and EDA (Crone et al., 2003; Desman et al., 2008)
involving executive function abilities, which may be improved by found no group differences between ADHD and controls on SCRs during
medication. In the other study (Karatekin et al., 2010), pupillary re- a go/no-go task under five different conditions of reinforcement
sponses (i.e., pupil dilations to visual stimuli) were equivalent in chil- (Desman et al., 2008) or in response to different levels of punishment
dren and adolescents with ADHD, during a pro- and anti-saccade task, (Crone et al., 2003). In these studies, SCRs were calculated over mul-
compared to typically developing controls. tiple trials, rather than locked to a specific stimulus and may have been
insensitive to more subtle fluctuations in skin conductance during the
3.2.1.4. Summary of cognitive task effects. Twenty-three studies task. Moreover, the emphasis was on inhibitory control, whereas in
analysed autonomic arousal during a cognitive task. Among these, 14 James et al. (2016), the primary focus of the task is on response se-
found signs of hypo-activation of the ANS, one study found signs of lection.
hyper-activation, while eight found no group differences and one In a study measuring the effects of rewards on time estimation/re-
reported mixed findings dependent on the autonomic measure. The production, Luman et al. (2008) found that children with ADHD did not
majority of studies reporting signs of hypo-activation of the ANS respond to reinforcement as controls did (i.e., by improving estimation
measured EDA. Summarising, the hypothesis that children with accuracy when given rewards). However, RTV and SCR amplitude were
ADHD have difficulties up-regulating autonomic arousal during normalised in response to reinforced feedback, compared to feedback-
cognitively challenging tasks, with consequent negative outcomes on only, in the ADHD group, suggesting that reinforcement enhanced ANS
behaviour and performance, is supported by the majority of the studies activity and improved time estimation in this group. One study
included in this review. However, as will be discussed in paragraph 4.5, (Wilbertz et al., 2013) investigating ANS activity during delay aversion,
the nature of the cognitive task or the presence of motivational factors an often-reported feature of ADHD (Van Dessel et al., 2018), found
(such as feedback; see paragraph 3.2.2), appears to influence the profile increased SCL (i.e., hyperarousal) during the delay interval in adults

Table 7
Summary of results from studies measuring pupil size measures during cognitive tasks.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect
hypo/hyper/none

Karatekin et al., 26 ADHD 48 typical; 29 Children and Pro- and Anti-saccade Pupil ADHD = TD controls and None
2010 psychosis adolescents task diameter psychosis
Wainstein et al., 28 22 Children and Visuo-spatial Working Pupil Off-medication ADHD < CTRL Hypo
2017 adolescents Memory task diameter On-medication ADHD = CTRL
Count (hypo:hyper:none) 1:0:1

194
 A. Bellato, et al.

Table 8
of results from studies measuring electrodermal activity during tasks involving reward processing.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect
Hypo/hyper/none

Beauchaine et al., 17 22 typical; Adolescents Choice RT + reinforcement ns-SCRs to extinction No group differences None (males only)
2001a 20 CD
Crone et al., 2003 22 22 Children and Hybrid go/no-go flanker task SCRs during task blocks No significant differences between None
adolescents with and without ADHD and controls on the increase of SCRs after punishment, positive
punishment. and negative feedback.
Desman et al., 2008 19 19 Children and Go-No Go task + reinforcement SCR No differences between groups None
(study 1) adolescents
Iaboni et al., 1997a 18 18 Children and Choice RT + reinforcement Mean SCL No differences between groups. Hypo (males only)
adolescents ADHD children showed no change in SCL between reward and

195
extinction (in controls, increase of SCL between reward and
extinction and decrease from extinction to reward).
James et al., 2016 73 + 75 72 + 72 Adolescents and 4-choice RT Fast Task + incentives SCL and SCRs ADHD < CTRl on SCL (only at slow-event rate with no incentives None
siblings siblings young adults and not during fast-event rate with incentive).
ADHD = TD on SCRs
Luman et al., 2008 25 30 Children and Time reproduction task SCR ADHD: reduced SCR in the feedback-only condition, compared with Hypo
adolescents reinforcement (effect not significant in CTRL)
Wilbertz et al., 2013 12 12 Adults Monetary incentive delay task SCL, SCRs ADHD > CTRl on mean SCL during delays. Hyper
ADHD = CTRl on SCRs after loss or gain.
Wilbertz et al., 2017 28 28 Adults Probabilistic anticipation of SCRs No differences between groups on SCR None
monetary loss & gain
COUNT (hypo:hyper:none) 2:1:5

a
Resting state measured before, during or after the task (see Tables 2–4 for resting state findings).
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
 A. Bellato, et al.

Table 9
Summary of results from studies measuring heart rate during tasks involving reward processing.
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group effect
Hypo/hyper/
none

Beauchaine et al., 17 22 typical; Adolescents Manual response RT task PEP, RSA ADHD + CD < CTRl on PEP reactivity to reward. Hypo (males
2001a 20 CD However, no significant differences between ADHD and only)
CTRL
Crone et al., 2003 22 22 Children and Hybrid go/no-go flanker task + feebdack Feedback-locked HR changes ADHD < CTRl on HR difference between positive and Hypo
adolescents negative feedback (HR was faster after positive feedback)
a
Crowell et al., 2006 18 ADHD + ODD 20 Pre-schoolers Recording of psychophysiological reactivity to PEP, RSA ADHD + ODD > TD on during-game PEP length Hyper
reward during playing with shape-based ADHD + ODD = TD on during-game RSA
commercially available game
Desman et al., 2008 19 19 Children and Go-No Go task + reinforcement Mean heart rate for each No significant differences between groups None (males
(study 1) adolescents reinforcement condition only)

196
Iaboni et al., 1997a 18 18 Children and Choice RT task – Reward and Extinction Mean HR More rapid habituation to reward in ADHD. Hypo (males
adolescents HR deceleration after extinction of rewards was reduced only)
in ADHD, compared to CTRL
Luman et al., 2007 18 18 Children and Time reproduction task HR acceleration and No differences between ADHD and CTRL on HR response Hypo
adolescents deceleration; feedback-locked to feedback.
IBI; HRV Unlike CTRL, no initial deceleration after feedback in
ADHD
Luman et al., 2008 25 30 Children and Time reproduction task HR acceleration and No differences in heart-rate responses to feedback, Hyper
adolescents deceleration; feedback-locked however faster reactivity to feedback in ADHD
IBI; HRV
Tenenbaum et al., 69 48 Children and Balloon Analog Risk Taking task Mean PEP and RSA ADHD and ADHD + ODD < CTRl on PEP lengthening Hypo
2018a adolescents and RSA increase (these were mostly explained by CD
and ODD symptoms).
Wilbertz et al., 2013 12 12 Adults Monetary incentive delay task Mean HR during delay ADHD > TD Hyper
COUNT (hypo:hyper:none) 5:3:1

a
Resting state measured before, during or after the reward task (see Tables 2–4 for resting state findings).
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

with ADHD, suggesting that stressful but engaging situations may over- increased reward sensitivity (measured using the Balloon Analogue
activate ANS functioning, an opposite reaction to low stimulation set- Risk-Taking Task), lengthening of PEP and reduced RSA withdrawal
tings such as resting or monotonous, simple cognitive tasks. In a later were explained by comorbid CD and ODD symptoms, and not by hy-
study (Wilbertz et al., 2017), the authors reported an increased neural peractivity or inattention, in a sample of children with ADHD. In ad-
response to monetary loss in adults with ADHD during a probabilistic dition to these studies, Crowell et al. (2006) found signs of autonomic
monetary loss or gain paradigm, but did not find any group differences hyper-activation both at baseline and in association with reward,
in SCR. The discrepant results are difficult to reconcile but suggest that during a tablet-based game played by pre-schoolers with
delay may be more aversive in ADHD, eliciting a more pronounced ANS ADHD + ODD. However, it is not fully clear if these effects were mainly
response. Alternatively, as the monetary loss paradigm was a passive driven by symptoms of ADHD, ODD or both.
task (requiring no response), the lack of SCR response may reflect the Lastly, Desman et al. (2008) reported no significant group differ-
different response requirements of the two paradigms. ences in HR (and EDA, described above) during a Go-No Go task: all
children (controls and ADHD) showed an increase in HR in the reward
3.2.2.2. Heart rate. Among the nine studies analysing the impact of and response cost conditions, compared to the neutral condition, but
positive or negative reward/reinforcement on autonomic measures of there were no group differences. It should be noted, therefore, that
HR functioning (Table 9), five reported hypo-activation of the ANS to measuring autonomic arousal at the block level (mean heart rate was
reward, three reported hyper-reactivity and one found no group used in this study) may have reduced sensitivity compared to mea-
difference between ADHD and controls. suring phasic changes, i.e., through the analysis of stimulus-locked
During a time production paradigm, Luman et al. (2007) found that variations in heart rate. Also, the medication washout period for some
children with ADHD generally showed HR deceleration following re- of the children on medication was less than 24 h, which may have been
sponse feedback, while controls showed an initial acceleration followed too short to remove the possibly normalising effects of medication on
by a deceleration, suggesting atypical modulation of ANS activity in ANS function.
ADHD when preparing to make a response following feedback. In ad-
dition, the authors reported evidence of reduced mental effort when 3.2.2.3. Summary of reward/reinforcement task effects. Of 11 studies
incentives were not provided to children with ADHD (reflected in re- assessing autonomic responses to reward or reinforcement in ADHD,
duced low frequency HRV in the reward and response cost conditions, three showed signs of hypo-activation of the ANS, two found hyper-
compared to neutral). A second study (Luman et al., 2008, described in activation, three did not find any group difference and three showed
paragraph 3.2.2.1), however, found different results, i.e., no group mixing results depending on the ANS measure. Group differences on
differences in HR or HR response to reinforcement and a trend for ANS reactivity to reward (hypo- or hyper-) were more likely when HR
ADHD children to show HR acceleration following reinforcement, measures were used, than EDA. Hypo-reactivity to reward was more
compared with controls, which would indicate ANS hyper-responsivity likely in tasks that measured extinction to reward or some aspect of
to reinforcement in ADHD. Similar results showing hyper-responsivity response control, whereas hyper-responsivity was more likely during
of ANS in response to reward, were also found by Wilbertz et al. (2013), time estimation and frustration, suggesting an emotional reaction
who found that adults with ADHD had increased heart rate during a during the task. However, the findings were inconsistent between
frustration-inducing monetary incentive task and during a delay period, studies and seem also to be influenced by comorbid symptoms of
and these effects were associated with self-reported impatience, ODD, as will be discussed in paragraph 4.5.
boredom and negative affect (and with increased SCL, see paragraph
3.2.2.1). 3.2.3. Socio-emotional tasks
Iaboni et al. (1997) found that children with ADHD showed an Ten studies measuring ANS functioning during a socio-emotional
equivalent increase in HR compared with controls, when rewards were task (Table 10) were included in this review because of the important
introduced during a motor response task, but the decrease in HR after role of emotional dysregulation and social impairments in ADHD. There
successive trials was more rapid in ADHD, which was interpreted as were too few studies to create specific categories for social and emo-
more rapid habituation to reward. Once rewards were extinguished, the tional tasks and so we have collapsed across them here to create a
control group showed the expected HR decrease but this was delayed in ‘socio-emotional’ category.
the ADHD group, suggesting hypo-responsivity to changes in re-
inforcement contingencies. This is partially in line with the EDA find- 3.2.3.1. EDA. Conzelmann et al. (2014), by using EDA measures, found
ings published in the same paper and, taken together, they suggest that signs of hypo-sensitivity to social information in children and
HR may be more sensitive, compared to EDA, to changes in an ex- adolescents with ADHD, when presented with a series of pictures of
perimental context where cognitive and reward processing mechanisms positive, negative or neutral valence. Within the ADHD group, children
are involved. Similarly, Crone et al. (2003) found signs of reduced who were unmedicated showed significantly reduced pre-stimulus SCL
changes in HR in response to different reinforcement contingencies in and reduced post-stimulus SCRs, compared with controls and
ADHD, suggesting that children with ADHD show a reduced ANS re- medicated ADHD children (see paragraphs 3.3 and 4.6 for a
sponse to reinforcers. discussion of medication effects). The other two studies investigating
Three studies explicitly investigated the influence of comorbidities EDA (McQuade and Breaux, 2017; McQuade et al., 2017) found no
on ANS function in ADHD. Beauchaine et al. (2001) designed a manual significant relationships between ADHD symptoms and SCRs during
response task, where male adolescents with ADHD and conduct dis- socio-emotional tasks (see below, paragraph 3.2.3.2, for results about
order pressed a button in response to a single-digit visual stimulus. HR).
Performance was initially rewarded before being extinguished, re-
instated and extinguished again. In addition to this, participants wat- 3.2.3.2. Heart rate. Among the eight studies which analysed HR
ched a video showing an escalating conflict between peers. Although measures during socio-emotional paradigms, three reported hypo-
the authors did not find any difference in task performance, reduced reactivity to the tasks, three reported increased autonomic reactivity,
autonomic reactivity to reward (reduced PEP reactivity) was related to while two did not find any differences between ADHD and controls
levels of aggression in ADHD/CD, compared to controls, suggesting (Table 10).
hypo-activation of SNS, but there was no significant difference between A number of studies in this area have attempted to induce emotions
ADHD-only and controls. Some limitations of this study are the small and measure ANS reactivity following emotion-induction. In one study,
sample size and the absence of female participants. Tenenbaum et al. Musser et al. (2018) measured autonomic arousal during an interaction
(2018) similarly focused on ADHD and CD/ODD, and found that between children/adolescents and their parents by using a frustration-

197
 A. Bellato, et al.

Table 10
Summary of results from studies measuring electrodermal activity and heart rate during socio-emotional tasks.
EDA measures

First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group
effect Hypo/
hyper/none
Conzelmann et al., 80 22 Children and Positive, negative and Mean SCL and ADHD-no-MPH < ADHD-with-MPH and TD on Baseline SCL. Hypo (males
2014b adolescents neutral valence task SCRs ADHD-no-MPH < ADHD-with-MPH and TD on SCRs to emotional pictures only)
McQuade and N/A 61 Low- vs High-ADHD Children and Social rejection SCL No association between SCL and ADHD traits. SCL to the tasks was associated None
Breaux, 2017a traits adolescents with internalising symptoms.
McQuade et al., N/A 61 Low- vs High-ADHD Children and Social rejection SCL No association between SCL and ADHD traits. None
2017a traits adolescents
COUNT (hypo:hyper:none) 1:0:2
HR measures
First author, year ADHD n Control n Age group Paradigm ANS measure Main findings ANS group
effect Hypo/
hyper/none
Bubier and N/A 63 TD boys, ADHD and Children Affective decision RSA, PEP Among males, reduced decision making skills were associated with symptoms Hypo (only
Drabick, 2008 ODD traits assessed but making and emotion- of hyperactivity and impulsivity, and this was mediated by reduced males)
not with clinical inducing task sympathetic activation in the emotion-inducing task
measures.
Lackschewitz et al., 18 18 Adults Tier Social Stress Test SDNN, RMSSD, ADHD < CTRl in the stress phase, and on HR increase during anticipation of Hypo
2008a Power Spectral stress
Density, LF, HF, ADHD > CTRl on decline from stress phase to the first post-stress phase
LF:HF

198
McQuade and N/A 61 Low- vs High-ADHD Children and Social rejection task RSA, RDA ADHD symptoms were associated with reduced RSA withdrawal in response Hyper
Breaux, 2017a traits adolescents reactivity to social rejection, after controlling for ODD and internalising symptoms
Musser et al., 2011a 32 32 Children and Emotion induction PEP, RSA No differences between ADHD and CTRL on PEP; Hypo
adolescents and suppression task ADHD < CTRl on increase of RSA from baseline to negative suppression
condition.
Positive induction condition: RSA increased from baseline, in ADHD, while it
decreased from baseline, in controls.
Musser et al., 2013a 21 ADHD-low-social-skills; 75 Children and Emotion induction PEP, RSA ADHD > CTRL > ADHD-low-social-skills on mean PEP length throughout Hyper
54 ADHD-high-social-skills adolescents and suppression task the task.
(ADHD = CTRL) > ADHD-low-social-skills on RSA increase from baseline to
induction and suppression of negative emotions.
ADHD-only showed RSA increase from baseline to positive induction
condition; CTRL & ADHD-low-social-skills: RSA decrease
Musser et al., 42 44 Children and Parent-Child PEP, RSA No differences between ADHD and CTRL on mean PEP. Hyper
2018a,c adolescents Interaction Task ADHD > CTRl on RSA withdrawal between baseline and task
Oliver et al., 2012a N/A 22 low-ADHD-traits; 20 Adults Driving simulation Heart period, RSA, No differences between low- and high-ADHD-traits None
high-ADHD traits task PEP
Waschbusch et al., 17 ADHD; 23 CD/ODD; 20 115 Children and Reaction to verbal Heart rate, event- No differences between ADHD and CTRL None
2002 ADHD + CD/ODD adolescents provocation task locked R-R
intervals
COUNT (hypo:hyper:none) 3:3:2

a
Resting state measured before, during or after the social task (see Tables 2–4 for resting state findings).
b
Measured effects of an intervention (see Table 11 for results of intervention).
c
An erratum for this paper (Musser et al., 2018) reported the same results found in the original.
Neuroscience and Biobehavioral Reviews 108 (2020) 182–206
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

inducing paradigm (Parent-Child Interaction Task). They found ex-

of SCL during resting-state, after caffeine administration, in ADHD this increase was not dependent on the

reduced baseline SCL during the socio-emotional task and SCRs in response to emotion-inducing pictures,

HR decelerations on error trials during a global/local task (selective attention) were found in controls, but
This study looked at the effects of caffeine, not medication. Although ADHD and TD showed similar increase

not in the unmedicated ADHD group. However, children with ADHD and under MPH, show normalised HR

Two bio-types identified based on cognitive performance. The impulsive biotype had a greater response to

MPH increased pupil size during the task, in ADHD. Pupil size was reduced between ADHD-off medication
No effect of MPH on EDA measures during resting at eyes-closed. During the auditory oddball task, MPH
maintained SCL constant throughout the task, in ADHD (up-regulation of arousal), while the decrease of

MPH normalised any group difference on SCL (ADHD < controls when unmedicated) and reduced SCR
cessive RSA withdrawal in the ADHD group when switching from

Children with ADHD under MPH were similar to TD, while unmedicated children with ADHD showed
baseline to the challenging task, reflecting PNS dysregulation and

and controls, and normalised (no significant difference) between ADHD-on medication and controls
hyper-reactivity to frustration. An earlier study by the same group
(Musser et al., 2011) measured induction or suppression of positive or
negative emotions and found, in the control group, an increase of RSA

ATX treatment (measured only as a reduction of symptoms), compared with inattentive.

from baseline to the negative-emotion suppression condition, reflecting

decelerations to errors and in response to feedback after errors, similarly to controls.

PNS withdrawal and a decrease in RSA from baseline to the positive-
emotion induction condition. These emotion-condition effects were
absent in ADHD, therefore the authors concluded that ADHD is asso-

(which was similar to CTRL when ADHD were unmedicated) during CPT.
SCL with task ongoing was larger in unmedicated ADHD, than controls.
ciated with PNS hypo-activation but intact sympathetic functioning (no
significant results were found for PEP). A further study (Musser et al.,
2013) used a similar paradigm and investigated whether co-occurring
callous and unemotional traits (CUT), measured as pro-social beha-
viours, influenced the findings in ADHD. Children with ADHD with
atypical pro-social behaviours showed reduced sympathetic function
(increased PEP) and PNS function (reduced RSA) throughout the entire
task compared to controls, while those with typical pro-social beha-
viours showed increased SNS activity (shorter PEP) throughout and also
compared to controls and ADHD + MPH.

significantly increased PNS activation during the positive induction

condition. These results suggest that ADHD children with typical pro-
social behaviours exhibit increased SNS and PNS activity during emo-
tional tasks, while those with low pro-social behaviours (indicative of
dose, while in TD it was.

the presence of CUT) show SNS and PNS hypo-arousal. This is con-
sistent with evidence of reduced ANS reactivity in those with conduct
disorder (Fanti et al., 2019).
Main findings

The association between socio-emotional abilities and cognitive

performance has been investigated by McQuade and Breaux (2017),
who analysed HR in a series of experimental paradigms, including an
executive function battery, aimed at inducing frustration. Overall, they
found that ADHD symptoms were correlated with reduced RSA with-
Resting-state and EDA measured during

Pupil size measured during a cognitive

Resting-state HR; HR during cognitive

drawal in response to social rejection, which would suggest a reduced

EDA during socio-emotional task

ability to down-regulate autonomic arousal in response to negative

emotions. Furthermore, children with poor EF showed better social and
EDA during cognitive task

academic skills if they demonstrated higher RSA withdrawal during the

HR during cognitive task
Summary of studies investigating the effect of ADHD medication or interventions on arousal measures.

frustration tasks, suggesting that the ability to efficiently regulate PNS

an attentional task

may be related to successful compensation of EF deficits, mitigating

Resting-state EDA

their impact on adaptive functioning. However, there was no direct

ANS measure

group comparison and the overall sample included only males, limiting
the generalisability of the study findings to females with ADHD. Fur-
task

task

thermore, the tasks did not tackle everyday failure experiences making
it difficult to interpret their relevance to general emotion regulation.
Lastly, among two studies investigating the impact of comorbid
symptoms on ANS functioning, Waschbusch et al. (2002) assessed re-
actions to verbal provocation during a competitive computer game in
Children and

Children and

Children and

Children and

Children and

Children and
Adolescents

boys with ADHD, ADHD + CD/ODD, CD/ODD. Although there were no

adolescents

adolescents

adolescents

adolescents

adolescents

adolescents
Age group

differences between pure ADHD and controls, the comorbid group

(ADHD + CD/ODD) showed a more pronounced ANS response to pro-
vocation (faster HR acceleration after losing points during game-play).
In another study (Bubier and Drabick, 2008), the association between
traits of ADHD and ODD, and HR measures, was investigated through
45 un-medicated + 35 with

16 un-medicated + 16 with

an affective decision making and emotion-inducing task in typically

developing children and adolescents. Among male participants, reduced
decision-making skills (lower proportion of advantageous compared to
prescribed MPH

prescribed MPH
Sample (ADHD)

disadvantageous choices) were associated with reduced sympathetic

activation (increased PEP length) during exposure to emotion-inducing
stimuli. Moreover, reduced sympathetic arousal during the socio-emo-
112

tional task was associated with traits of hyperactivity and impulsivity,

18

34

18

28

but not ODD, indicating an association between traits of hyperactivity/

Conzelmann et al., 2014

impulsivity, SNS dysfunctioning and decision-making skills.

Hermens et al., 2005b

Wainstein et al., 2017

Lawrence et al., 2005

Leikauf et al., 2017

Groen et al., 2009
Barry et al., 2012
First author, year

3.2.3.3. Summary of socio-emotional task effects. Nine studies

investigated autonomic arousal in ADHD during a socio-emotional
task. Among these, four found signs of hypo-reactivity during the
Table 11

tasks, two found signs of hyper-reactivity and two did not find any
group differences. The studies targeted different aspects of social

199
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

information and emotion processing making it difficult to draw firm attention and response regulation, and a greater likelihood of hypoar-
conclusions about the profile of ANS regulation in ADHD in relation to ousal in reward tasks in studies using HR than EDA measures.
socio-emotional processing, as later discussed in paragraph 4.5. The greater predominance of effects of hypo-arousal compared with
hyper-arousal, in studies that reported a group difference, suggests re-
3.3. Effects of medication and other interventions duced autonomic arousal in ADHD which is in line with theoretical
models proposing arousal (dys)regulation as an important factor in the
Six studies investigated the effects of ADHD medication, e.g., me- pathology of ADHD (Geissler et al., 2014; Kuntsi and Klein, 2012; Sara
thylphenidate (MPH), five of which found signs of normalisation of ANS and Bouret, 2012; Sergeant, 2000). The review enabled us to address
functioning (Table 11). Two studies (Hermens et al., 2005b; Leikauf the three main questions set out in the introduction. Firstly, we found
et al., 2017) measured autonomic arousal at rest and did not find any evidence of ANS dysfunction in ADHD (question 1) either in the di-
effect of MPH on EDA (Hermens et al., 2005b) or HR measures (Leikauf rection of hypo- or hyper-arousal, in 52 of 91 findings reported across
et al., 2017). Most of the other studies have instead investigated the the 55 studies included in this review. Secondly, most of the significant
effects of stimulant medication during cognitive (Groen et al., 2009; group differences were for hypo-activation of the ANS in ADHD rather
Lawrence et al., 2005; Leikauf et al., 2017), socio-emotional than hyper-activation (question 2). Lastly, our review identified find-
(Conzelmann et al., 2014) or attentional tasks (Hermens et al., 2005b; ings of ANS dysfunction in ADHD during resting-state and during ex-
Wainstein et al., 2017). Conzelmann et al. (2014), for example, showed perimental tasks requiring the processing of cognitive, rewarding or
that unmedicated children with ADHD had reduced baseline SCL during socio-emotional information, indicating reduced functioning of the ANS
a socio-emotional task and reduced SCRs in response to emotion-in- at rest and impaired adaptation of arousal in response to task demands
ducing pictures, while there was no significant difference between (question 3). We now consider these findings in more detail.
ADHD on-medication and controls. Groen et al. (2009) measured the
effect of MPH on HR measures during a cognitive task, and found HR 4.2. Resting state findings
decelerations to errors and in response to feedback after errors in both
children with ADHD on MPH and controls, while these decelerations The evidence of atypical functioning of the ANS during rest in-
were not present when the ADHD children were unmedicated. Simi- dicates that insufficient resources are allocated to spontaneous neural
larly, autonomic measures during an auditory oddball task have been activity in situations where attention is not focused on a single task or
collected by Hermens et al. (2005b), who showed that MPH maintained event. Among the studies reporting a group difference at rest, more
SCL at a more constant level throughout the task, in ADHD, while a reported hypo- than hyper-arousal and this was particularly clear for
decrease in SCL, over time, was present in unmedicated children with studies using EDA. Fewer studies using HR measures reported a group
ADHD. Using SCL, Lawrence et al. (2005) found similar results during difference and the findings from pupillometry studies were incon-
an attentional task, so that children with ADHD, when unmedicated, clusive. Since EDA is thought to be a more pure measure of SNS activity,
showed reduced SCL compared to controls, but this difference was not than PNS, this pattern of findings suggests under-activation of the SNS
evident when they were medicated. The only study using pupillometry at rest in ADHD. This is consistent with evidence of abnormal resting
(Wainstein et al., 2017) found increased pupil size during a cognitive state brain activity in ADHD, including increased slow-wave EEG power
task, when children and adolescents with ADHD were on medication, (Barry et al., 2003, 2009) and reduced functional connectivity within
compared to when they were not. and between resting-state brain networks, including the DMN (Rubia,
One other study focused on analysing the effects of caffeine (Barry 2018).
et al., 2012) and found that in controls and children/adolescents with It is important to emphasise that a number of studies included in
ADHD, caffeine increased SCL at rest. However, while in controls this this review reported no group differences on any ANS measure during
increase was dependent on the dose, in ADHD it was not, suggesting resting state. These heterogeneous findings undermine the theory that
reduced sensitivity to the dose of caffeine, which is a natural stimulant, arousal is impaired in ADHD. Interestingly, more recent evidence from
in ADHD. Since this was the only study investigating caffeine, a clear EEG studies also calls into question the reliability of the findings of
conclusion cannot be fully supported. increased slow relative to fast EEG waves in ADHD (see Saad et al
Summarising, results from these studies support the hypothesis of (2015) and Loo and Makeig (2012) for reviews) indicating that ANS and
hypo-arousal in ADHD, since most of them showed some up-regulating CNS arousal may not be consistently impaired in ADHD. Although
or normalising effects of stimulant medication on measures of ANS much effort has been invested in trying to validate specific EEG bio-
functioning, in ADHD, which will be further discussed in paragraph 4.3. markers of ADHD, such as increased theta-beta ratio (TBR), studies
have been inconsistent in showing a clear association between TBR and
4. Discussion ADHD diagnosis (Saad et al., 2015). A recent review of the literature by
Newson and Thiagarajan (2018) revealed that atypical EEG profiles,
4.1. Summary and interpretation of main findings such as increased power at lower frequencies and decreased power at
higher frequencies, seem to be shared by individuals with different
We conducted a systematic review of the literature to identify stu- conditions (e.g., ADHD, OCD and schizophrenia) and not specific to
dies which measured ANS functioning in ADHD at rest and during tasks ADHD. Alternatively, there may be sub-groups within the ADHD po-
of cognition, reward/reinforcement and socio-emotional processing. pulation who are more or less likely to be characterised by atypical
Overall, our review of 55 studies produced 91 findings comparing ANS arousal, but equally plausible is the possibility that atypical ANS is
measures in people with traits of ADHD (clinical or subclinical) against driven by other factors commonly associated with ADHD, rather than
typical controls (see SM3 for a summary). The findings were hetero- ADHD itself.
geneous: 39 of 91 findings (43 %) were null, 12 of 91 (13 %) were in the Most studies that investigated the relationships between measures
direction of hyper-arousal in ADHD and 40 of 91(44 %) were in the of ANS functioning, such as changes in heart rate and HRV, electro-
direction of hypo-arousal in ADHD. Although mixed, the pattern overall dermal activity and pupil size, and the CNS, focused on sleep or on the
suggests that findings of either hypo-arousal or no group difference sleep-wakefulness transition. A review of the literature by de Zambotti
were more prominent than hyper-arousal in these studies. Moreover, et al. (2018), for example, showed that CNS and ANS measures couple
the effects were partly task- and measure-dependent with a greater during sleep, so that brain oscillatory activity parallels fluctuations in
likelihood of hypoarousal in resting state studies measuring EDA than cardiac activity. Huang et al. (2018) instead focused on the period
those measuring pupil size or HR in the resting state, a greater tendency before sleep onset and found that a reduction of activity of the ANS
for hypo- than hyperarousal when cognitive tasks required sustained (reflected in reduced mean HR and SCL over time) paralleled the

200
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

decrease in vigilance and wakefulness before sleep onset, calculated such as CD/ODD with which ADHD is highly comorbid, it will be im-
through an EEG-based algorithm. Conversely, Barry et al. (2005, 2007, portant to explore the extent to which the brain systems supporting
2008) investigated the relationships between EEG spectral power and these links between ANS and socio-emotional processing are affected in
SCL during resting-state and found that increased alpha power was these conditions.
associated with reduced SCL during eyes-closed resting-state, while It is important to point out that many of the effects from studies
switching to eyes-open caused an increase in SCL and a decrease in using cognitive, reward/reinforcement and socio-emotional tasks re-
alpha power. These findings suggests that specific measures of CNS and ported in this review were null effects. It proved difficult to identify
ANS activity are likely to change in parallel between different states of methodological parameters that might be more likely to elicit hy-
vigilance and, therefore, they may similarly reflect different states of poactivation of the ANS compared with no group differences (see
arousal. Section 4.5 below for further consideration of methodological factors).
The predominance of null findings is at odds with theories of arousal
4.3. Findings from studies using cognitive, reward/reinforcement or socio- dysregulation in ADHD which propose that such dysregulation con-
emotional tasks tributes to cognitive and behavioural features of the condition. The null
findings may be driven by task design, measurement error or small
The evidence of hypo-activation of the ANS during cognitive tasks sample sizes. Further research is needed to clarify the role of ANS in
described in this review is consistent with previous findings of greater cognition more broadly and in ADHD to determine whether it is more
RTV (Kofler et al., 2013) and response errors in ADHD, which are ex- important in some cognitive functions than others, or whether it is
acerbated when task stimuli are presented at a slower event rate (Metin simply a correlate of these cognitive features but not causally related to
et al., 2012). This suggests dysfunction in the brain systems linking ANS them.
activity with cognitive functioning (Aston-Jones and Cohen, 2005; Sara
and Bouret, 2012), particularly when upregulation of ANS activity is 4.4. The findings of the review in relation to models of arousal
required to support performance during long or monotonous tasks. This
is consistent with models of ADHD which propose that arousal reg- The findings of this review may be usefully considered in the con-
ulation difficulties represent a core pathway or mechanism underlying text of broader models of the relationships between ANS and behaviour.
the cognitive and behavioural features of ADHD (Geissler et al., 2014; The maintenance of a task-focussed state is dependent upon adjust-
Kuntsi and Klein, 2012; Sergeant, 2005). The findings of ANS hypoar- ments of arousal during a task. Animal studies investigating the LC-NE
ousal are also consistent with evidence of poorer performance when system (Aston-Jones and Cohen, 2005) have shown that an initial
cognitive tasks have a slow event rate (Metin et al., 2012; Wiersema phasic LC-NE response to a novel or salient stimulus is usually ex-
et al., 2006) and suggest that lower ANS activity may be one factor that tinguished quite rapidly if the stimulus proves not to be sufficiently
contributes to poorer performance of people with ADHD when tasks rewarding, so that the phasic LC-NE response habituates and ex-
require sustained attention and response regulation. It has also been ploratory non-task-focussed behaviours, e.g., low vigilance, emerge.
suggested that a general state of hypo-arousal in ADHD may be com- Conversely, if the stimulus is evaluated as rewarding or salient, the
pensated through maladaptive strategies such as hyperactive motor phasic mode is maintained through frequent firing bursts of LC-NE
behaviours and sensation seeking. These auto-regulatory strategies are neurons and release of NE, supporting sustained attention over time
therefore used to stabilize vigilance by creating a stimulating environ- (Gompf et al., 2010). The evaluation of costs and benefits associated to
ment. This is likely to be especially true in situations where regulation a stimulus or a situation, mainly involves the vmPFC, Orbito-Frontal
of arousal is more difficult, for example during mentally challenging or Cortex (OFC) and the Anterior Cingulate Cortex (ACC) (Aston-Jones
boring situations, but it may vary significantly from person to person and Cohen, 2005), which have a top-down regulatory role on LC and
and according to the level of general tiredness. In support of this in- send efferent signals to maintain or interrupt the activation of LC-NE
terpretation, the tasks most likely to elicit autonomic hypoarousal in system in the phasic mode. Previous research has consistently shown
ADHD were tasks of attention requiring an infrequent response to a rare evidence of impaired maintenance of attention in ADHD (Cubillo et al.,
target or tasks of response control, rather than tasks requiring complex 2012) and the evidence presented in this review suggests that ANS
executive functions such as planning or working memory. hypo-activity may be an important component of this dysfunction,
The review also identified atypical ANS response to rewards and potentially contributing to a failure of the LC-NE system to maintain a
reinforcers. There was a predominance of findings of hypoarousal in task-focussed state. Moreover, studies included in this review indicate
studies using HR measures but studies using EDA did not show con- that impaired error processing reportedly previously in ADHD (Shiels
sistent results with many studies reporting no group differences. The and Hawk, 2010) is reflected in weaker ANS responses to errors. One
findings from the HR studies are consistent with research showing re- possibility requiring further investigation is that hypo-activation of the
duced responsivity to rewards in ADHD and with broader models which ACC in response to errors previously reported in ADHD (Fallgatter
hypothesise that atypical reward processing represents a core deficit in et al., 2004; Liotti et al., 2005; Rubia et al., 2019) may undermine the
ADHD (Sonuga-Barke, 2002; Sonuga-Barke and Halperin, 2010). In- phasic LC-NE response and the consequent upregulation of arousal to
terestingly, ANS hyper-responsivity to rewards was reported in some increase attention and reduce the chance of further errors. The estab-
studies that combined rewards with performance of a cognitive task, so lished links between ACC, vmPFC and the ANS (Critchley and Garfinkel,
that the introduction of rewards increased ANS functioning and im- 2018) suggest that impairments in error processing, ACC hypo-activa-
proved performance in ADHD. This indicates that impaired arousal tion and reduced ANS responsivity to errors may be crucially linked in
regulation in ADHD may be malleable by motivational incentives, ADHD. Further research is needed to investigate more closely the links
consistent with previous evidence showing the effects of motivational between ANS activity and the neural systems supporting cognition in
incentives on electrophysiological (Groom et al., 2010, 2013) and fMRI ADHD.
correlates (Liddle et al., 2011) in ADHD. However, this review also Besides the cortical and subcortical systems responsible for atten-
highlights the importance of considering the way in which rewards are tional and cognitive processes, other neural pathways and brains
delivered; several studies showed reduced sensitivity to reinforcers in structures, e.g., the amygdala and limbic regions, have been shown to
ADHD, indicating that the timing, nature and size of these may need to be involved in processing of rewards and socio-emotional stimuli
be optimised. (Murray, 2007). Moreover, an association between autonomic me-
The findings from studies using socio-emotional tasks were highly chanisms, brain and emotional responses (such as facial expressions)
heterogeneous and too few in number to draw strong conclusions. has been proposed by the polyvagal theory (Porges, 2009), which
Given that these behaviours are characteristically atypical in disorders suggests that atypical functioning of different ANS mechanisms may be

201
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

linked to difficulties in regulating socio-emotional and communication controlled for the presence of these symptoms (see Table 12 in sup-
behaviours. The findings of this review indicate that the ANS response plementary Materials, SM2, for a summary). About 22 studies (out of
to these socio-emotional contexts are atypical in ADHD but the relative 56) excluded individuals with ADHD and comorbid ODD or CD, while
lack of research in this area precludes a detailed understanding of the symptoms of ASD were not explicitly measured in 33 studies and
role of the ANS in this aspect of ADHD. children with ASD were excluded from 22 studies. Further research is
therefore needed to enlighten the mechanisms underlying the interac-
4.5. Methodological factors in task design, data analysis, sample size and tion between ANS functioning, ADHD and comorbid symptoms, bearing
characteristics in mind that excluding children with comorbid symptoms may reduce
the impact of potential confounding factors but narrow the possibility
To obtain a clearer picture of ANS functioning in ADHD, in this of generalising any results. Moreover, there may be important avenues
review, we investigated whether specific methodological choices made for the identification of sub-groups based on biomarkers such as ANS
during task design and analyses could explain the heterogeneity of re- activity that cut across traditional diagnostic boundaries and that lead
sults. Traditional research settings, e.g., fMRI or EEG laboratories, to alternative ways of clustering subgroups of patients based on similar
provide a naturally stimulating environment that may increase arousal, profiles (Collins et al., 2011).
although it was difficult to determine whether this affected the studies Studies focusing on children with subthreshold traits of ADHD or
reviewed here. Moreover, although it could be argued that arousal at analysing the association between continuous levels of symptoms and
rest may depend on whether a task of some sort is expected to follow arousal measures have been included in this review, however in most of
on, resting-state autonomic dysfunctions were found in ADHD irre- the cases the results were not significant. There were very few studies
spective of whether a cognitive task was presented before or after. taking this approach and of those included here, most did not include
Similarly, the length of the resting-state period was found not to impact participants from across the spectrum, tending instead to recruit par-
ANS functioning, with studies equally likely to report hypo- or hyper- ticipants with a range of scores at the typical end of the dimension. It is
arousal with both short and long resting periods. too early to say therefore whether ANS dysfunction is related to ADHD
When evaluating methodological choices regarding the analyses of as a continuous trait measure. Further research is needed to address this
measures of ANS functioning, studies which analysed measures at question. Including broader samples of children with different pre-
block-level (e.g., averaged over a long period of time, rather than sti- sentations of ADHD symptoms (e.g., inattentive-only, hyperactive-only
mulus-locked measurements), were less likely to report group effects, and combined), clinical and sub-threshold symptoms; and comorbid
suggesting that this approach may not be sufficiently sensitive to detect symptoms may better represent the vast clinical heterogeneity of ADHD
phasic variations of autonomic functioning. Besides this, the most symptomatology and inform about ANS and cognitive functioning in
consistent evidence of hypo-arousal at rest came from studies mea- children with different symptom profiles.
suring EDA, which is thought to be a pure measure of SNS activity, since
the sweat glands are primarily innervated by adrenergic receptors 4.6. Effects of medication
(Berntson et al., 1997). Even when using cognitive tasks, signs of hypo-
arousal were predominantly found in ADHD through EDA, with only Some of the studies included in our review focussed on analysing
one study showing hyper-arousal. Although HR may be less sensitive to the impact of medication on ANS functioning. Stimulant medication for
detect primary deficits of SNS or PNS at rest, signs of hypo-arousal were ADHD, e.g., MPH, is usually a first-choice treatment for ADHD (NICE,
found in ADHD during cognitive tasks when HR was measured con- 2018) but, together with well documented findings of improvement in
currently. Tasks which were more likely to be associated with arousal cognitive performance (Coghill et al., 2014) and reduction of hyper-
dysregulation were those involving sustained attention, error mon- activity and impulsivity (Faraone and Buitelaar, 2010), it also has an
itoring and response inhibition, especially if presented at a slow pace, effect on ANS functioning, for example increasing HR and blood pres-
while null effects seem to emerge from more challenging paradigms, sure (Liang et al., 2018) and giving rise to difficulties in sleeping
e.g., involving executive functions or fast paced tasks. (Storebo et al., 2015).
Studies which analysed autonomic arousal during the processing of Results from the studies included in this review are in line with the
social information were very different from one another, making it hypo-arousal theory of ADHD, and indicated that stimulant medication,
difficult to interpret the results emerging from the present review. It e.g., MPH, may up-regulate or normalise a general state of hypo-arousal
could be that some effects, e.g., hyper-reactivity, were mainly driven by in ADHD. However, studies were divergent when fully reviewed (e.g.,
the challenging nature of the situation and not by its social features. If planning different washout periods, taking different decisions about
this were the case, signs of hypo-arousal found during socio-emotional leaving some children on medication during the testing, within-sample
tasks may resemble difficulties in allocating energetic resources ac- differences in medication history), therefore, further research is needed
cording to the context of the task, instead of specifically reflecting to investigate the relation between effects of ADHD medications on
hypo-sensitivity to social stimuli. Since emotional dysregulation and cognitive, behavioural and arousal mechanisms. Moreover, there were
atypical social abilities are key behavioural features of ADHD, it will be no studies investigating the effect of non-stimulant medication such as
important to focus on this domain in future research and develop atomoxetine, clonidine or guanfacine. Non-stimulants, particularly
paradigms that challenge these processes more precisely. clonidine and guanfacine, are likely to have an opposite effect on ANS
Other methodological issues found in some of the studies were re- functioning, i.e., reducing HR and blood pressure (see, for example,
garding the sample characteristics. For example, although the sample Sayer et al., 2016) compared with stimulants, but they have similar
size of studies included in this review was variable (10 to almost 500), benefits on cognitive performance and ADHD symptoms. Further re-
there was no clear relationship between sample size and the presence or search directly comparing the effects of stimulants and non-stimulants
directionality of effects on ANS measures. Moreover, 14 studies only could further illuminate the relationships between ANS activity and
included single-gendered participants (13 only males and 1 only fe- cognition in ADHD and may also help identify sub-groups, given that
males), giving rise to evident issues of generalisability to the ADHD individuals with ADHD are often more responsive to one type of med-
population. ication than the other.
Another important but overlooked aspect of research into ADHD is
the influence of comorbid symptoms. Although conduct (CD), opposi- 4.7. Implications, limitations and future directions
tional-defiant (ODD) and Autism Spectrum (ASD) disorders, are some of
the most frequently reported comorbidities in ADHD (Jensen and Although recent developments in imaging technology are improving
Steinhausen, 2015), only few of these studies explained how they localisation and functional imaging of the LC and its connections with

202
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

other regions, the small size and deep location of this structure make it We conclude our review by indicating some interesting areas of
hard to image, even in adults. Therefore, measuring other indices of LC- research which we think should be targeted in future studies, in order to
NE involvement, such as measures of ANS functioning at rest and obtain more knowledge about the relation between behaviour, cogni-
during cognitive tasks, may give some insights into the functioning of tion, attention and autonomic arousal mechanisms in ADHD:
the LC at any age and in neurodevelopmental populations, including
ADHD. Previous findings suggesting that specific measures of CNS and 1 Measuring different presentations of ADHD symptoms, both quali-
ANS activity are likely to change in parallel between different states of tatively (e.g., comparing inattentive-, hyperactive- and combined-
vigilance (e.g., de Zambotti et al., 2018; Huang et al., 2018; Barry et al., presentations of ADHD) and quantitatively (e.g., comparing children
2005, 2007, 2008) should be followed up by further research to better with different levels of impairment and adaptive functioning), to
explain the relationships between measures of CNS activity, such as verify the presence of similarities and differences on their auto-
EEG, and the ANS. Therefore, future studies should focus on carefully nomic arousal profiles;
designing experimental situations where autonomic (e.g., HR, EDA and 2 Measuring co-morbid symptoms in ADHD, such as ASD or OCD/
pupillometry), neuroimaging (e.g., EEG or fMRI) and behavioural ODD, instead of excluding individuals with comorbid conditions, to
measures are collected, during periods of resting-state and cognitive or identify profiles of autonomic functioning in different symptom
attentional tasks, both in ADHD and neurotypical individuals. profiles;
Knowing the associations between specific signs of ANS dysfunction 3 Assessing the effects of age and gender on ANS activity under dif-
in individuals with different symptomatological profiles, may help ferent paradigms;
clinicians in making more reliable and objective diagnoses, e.g., in the 4 Investigating the specific effects of medical and non-medical inter-
case of children with ADHD and comorbid ASD, or CD/ODD. Moreover, vention on short- and long-term changes in autonomic arousal in
increasing our knowledge about the effects of different medical inter- individuals with ADHD, and measuring the positive or negative
ventions on ANS functioning, may be helpful to study the response to impact of different types of information (e.g., reward or feedback)
stimulant and non-stimulant ADHD medication and try to develop in- on arousal regulation;
novative treatments which benefit both ANS and cognitive functioning 5 Investigating underlying mechanisms of socio-emotional processing
and limit the undesirable side effects. in ADHD, at neural, behavioural and electrophysiological level, and
This review also suggests that manipulating the nature of rewards evaluate the impact of co-occurring symptoms of CD and ASD on
and reinforcers provided to individuals with ADHD may help them to these mechanisms;
up-regulate their attentional and arousal state, with evident benefits in 6 Combining neuroimaging, electrophysiological and behavioural
many settings, such as schools or workplaces, and positive effects on measures in different experimental situations, such as cognitive-at-
productivity and academic achievement, quality of life and social in- tentional tasks and resting-state, and use them to derive subtypes
teractions. Equally, this could be applied to the home environment. based on physiological arousal patterns, cognitive function and
However, further research in this field is needed to develop innovative symptom profile.
programs for parents, teachers and employers, who can implement
strategies to optimise arousal regulation. In many cases, medication Acknowledgements
may still be the preferred first-line treatment. It is therefore imperative
in future research that the modes of action of currently available ADHD This work was supported by The Waterloo Foundation [grant
medications are better understood. In this review, we only found papers number 980-365], the National Institute for Health Research
assessing the effects of stimulants on ANS activity. Further research Nottingham Biomedical Research Centre Mental Health & Technology
assessing the effects of non-stimulants (e.g. atomoxetine, guanfacine, Theme [grant number BRC-1215-20003] and the University of
clonidine) on ANS activity and on the LC-NE system is needed parti- Nottingham.
cularly as these medications are often more effective in individuals who
do not respond to stimulants, lending weight to the possibility of sub- Appendix A. Supplementary data
types within ADHD, and they target the LC-NE system (Minzenberg
et al., 2008). Similarly, there are non-pharmacological interventions Supplementary material related to this article can be found, in the
under development that are also likely to impact an arousal regulation online version, at doi:https://doi.org/10.1016/j.neubiorev.2019.11.
deficit such as exercise interventions (Ng et al., 2017) and neurofeed- 001.
back (Arns et al., 2014). Reviewing the evidence on each of these in-
terventions is beyond the scope of this review but it will be important to References
test the effects of these interventions on autonomic and central nervous
system markers of arousal. American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental
Although the results of this review offer some useful factors to Disorders: DSM-5. https://doi.org/10.1176/appi.books.9780890425596.
Arns, M., Heinrich, H., Strehl, U., 2014. Evaluation of neurofeedback in ADHD: the long
consider in the design of future studies, there are some limitations to and winding road. Biol. Psychol. 95, 108–115. https://doi.org/10.1016/j.biopsycho.
address. First of all, the vast heterogeneity in the design of the studies 2013.11.013.
and/or in the methodology of data collection prevented us from car- Aston-Jones, G., Rajkowski, J., Cohen, J., 2000. Locus coeruleus and regulation of be-
havioral flexibility and attention. Prog. Brain Res. 126, 165–182. https://doi.org/10.
rying out a meta-analysis. There is also likely to be a bias in the in- 1016/S0079-6123(00)26013-5.
clusion of studies in that we only selected peer-reviewed publications. Aston-Jones, G., Cohen, J.D., 2005. An integrative theory of locus coeruleus-nor-
This increases confidence in the reliability of the findings from in- epinephrine function: adaptive gain and optimal performance. Annu. Rev. Neurosci.
28, 403–450. https://doi.org/10.1146/annurev.neuro.28.061604.135709.
dividual studies included in this review but there may be unpublished
Barry, R.J., Clarke, A.R., Johnstone, S.J., 2003. A review of electrophysiology in atten-
findings that would give valuable information. Since this is likely to be tion-deficit/hyperactivity disorder: I. Qualitative and quantitative electro-
a common limitation of most review studies, we encourage open sci- encephalography. Clin. Neurophysiol. 114, 171–183. https://doi.org/10.1016/
S1388-2457(02)00362-0.
ence and pre-registered studies with clear hypotheses to facilitate the
Barry, R.J., Rushby, J.A., Wallace, M.J., Clarke, A.R., Johnstone, S.J., Zlojutro, I., 2005.
inclusion of unpublished analyses, (including null findings) in future Caffeine effects on resting-state arousal. Clin. Neurophysiol. 116, 2693–2700.
literature reviews. Moreover, since we could not properly investigate https://doi.org/10.1016/j.clinph.2005.08.008.
the potential effects of sex and age on ANS activity in ADHD due to the Barry, R.J., Clarke, A.R., Johnstone, S.J., Magee, C.A., Rushby, J.A., 2007. EEG differ-
ences between eyes-closed and eyes-open resting conditions. Clin. Neurophysiol. 118,
vast heterogeneity of samples characteristics and/or missing informa- 2765–2773. https://doi.org/10.1016/j.clinph.2007.07.028.
tion in the published full texts, we strongly advise that these char- Barry, R.J., Clarke, A.R., Johnstone, S.J., Rushby, J.A., 2008. Timing of caffeine's impact
acteristics are made a focus of future investigations in this area. on autonomic and central nervous system measures: Clarification of arousal effects.

203
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Biol. Psychol. 77, 304–316. https://doi.org/10.1016/j.biopsycho.2007.11.002. Dupuy, F.E., Clarke, A.R., Barry, R.J., Selikowitz, M., McCarthy, R., 2014. EEG and
Barry, R.J., Clarke, A.R., Johnstone, S.J., McCarthy, R., Selikowitz, M., 2009. electrodermal activity in girls with Attention-Deficit/Hyperactivity Disorder. Clin.
Electroencephalogram theta/beta ratio and arousal in attention-deficit/hyperactivity Neurophysiol. 125, 491–499. https://doi.org/10.1016/j.clinph.2013.09.007.
disorder: evidence of independent processes. Biol. Psychiatry 66, 398–401. https:// Dykman, R.A., Ackerman, P.T., Oglesby, D.M., Holcomb, P.J., 1982. Autonomic re-
doi.org/10.1016/j.biopsych.2009.04.027. sponsivity during visual search of hyperactive and reading-disabled children. Pavlov.
Barry, R.J., Clarke, A.R., McCarthy, R., Selikowitz, M., MacDonald, B., Dupuy, F.E., 2012. J. Biol. Sci. 17, 150–157. https://www.ncbi.nlm.nih.gov/pubmed/7133780.
Caffeine effects on resting-state electrodermal levels in AD/HD suggest an anomalous Erskine, H.E., Norman, R.E., Ferrari, A.J., Chan, G.C., Copeland, W.E., Whiteford, H.A.,
arousal mechanism. Biol. Psychol. 89, 606–608. https://doi.org/10.1016/j. Scott, J.G., 2016. Long-term outcomes of Attention-Deficit/Hyperactivity disorder
biopsycho.2012.01.004. and conduct disorder: a systematic review and meta-analysis. J. Am. Acad. Child
Bast, N., Poustka, L., Freitag, C.M., 2018. The locus coeruleus-norepinephrine system as Adolesc. Psychiatry 55, 841–850. https://doi.org/10.1016/j.jaac.2016.06.016.
pacemaker of attention - a developmental mechanism of derailed attentional function Fallgatter, A.J., Ehlis, A.C., Seifert, J., Strik, W.K., Scheuerpflug, P., Zillessen, K.E.,
in autism spectrum disorder. Eur. J. Neurosci. 47, 115–125. https://doi.org/10.1111/ Herrmann, M.J., Warnke, A., 2004. Altered response control and anterior cingulate
ejn.13795. function in attention-deficit/hyperactivity disorder boys. Clin. Neurophysiol. 115,
Beauchaine, T.P., Katkin, E.S., Strassberg, Z., Snarr, J., 2001. Disinhibitory psycho- 973–981. https://doi.org/10.1016/j.clinph.2003.11.036.
pathology in male adolescents: discriminating conduct disorder from attention-def- Fanti, K.A., Eisenbarth, H., Goble, P., Demetriou, C., Kyranides, M.N., Goodwin, D.,
icit/hyperactivity disorder through concurrent assessment of multiple autonomic Zhang, J., Bobak, B., Cortese, S., 2019. Psychophysiological activity and reactivity in
states. J. Abnorm. Psychol. 110, 610–624. https://doi.org/10.1037//0021-843x.110. children and adolescents with conduct problems: a systematic review and meta-
4.610. analysis. Neurosci. Biobehav. Rev. 100, 98–107. https://doi.org/10.1016/j.
Berntson, G.G., Bigger Jr., J.T., Eckberg, D.L., Grossman, P., Kaufmann, P.G., Malik, M., neubiorev.2019.02.016.
Nagaraja, H.N., Porges, S.W., Saul, J.P., Stone, P.H., van der Molen, M.W., 1997. Faraone, S.V., Buitelaar, J., 2010. Comparing the efficacy of stimulants for ADHD in
Heart rate variability: origins, methods, and interpretive caveats. Psychophysiology children and adolescents using meta-analysis. Eur. Child Adolesc. Psychiatry 19,
34, 623–648. https://doi.org/10.1111/j.1469-8986.1997.tb02140.x. 353–364. https://doi.org/10.1007/s00787-009-0054-3.
Borger, N., Van Der Meere, J., Ronner, A., Alberts, E., Geuze, R., Bogte, H., 1999. Heart Faraone, S.V., Rostain, A.L., Blader, J., Busch, B., Childress, A.C., Connor, D.F., Newcorn,
rate variability and sustained attention in ADHD children. J. Abnorm. Child Psychol. J.H., 2019. Practitioner Review: emotional dysregulation in attention-deficit/hyper-
27, 25–33. https://doi.org/10.1023/A:1022610306984. activity disorder - implications for clinical recognition and intervention. J. Child
Borger, N., van der Meere, J., 2000. Motor control and state regulation in children with Psychol. Psychiatry 60, 133–150. https://doi.org/10.1111/jcpp.12899.
ADHD: a cardiac response study. Biol. Psychol. 51, 247–267. https://doi.org/10. Fayyad, J., De Graaf, R., Kessler, R., Alonso, J., Angermeyer, M., Demyttenaere, K., De
1016/S0301-0511(99)00040-X. Girolamo, G., Haro, J.M., Karam, E.G., Lara, C., Lepine, J.P., Ormel, J., Posada-Villa,
Brown, J.W., Alexander, W.H., 2017. Foraging value, risk avoidance, and multiple control J., Zaslavsky, A.M., Jin, R., 2007. Cross-national prevalence and correlates of adult
signals: how the anterior cingulate cortex controls value-based decision-making. J. attention-deficit hyperactivity disorder. Br. J. Psychiatry 190, 402–409. https://doi.
Cogn. Neurosci. 29, 1656–1673. https://doi.org/10.1162/jocn_a_01140. org/10.1192/bjp.bp.106.034389.
Bubier, J.L., Drabick, D.A.G., 2008. Affective decision-making and externalizing beha- Frazier, T.W., Demaree, H.A., Youngstrom, E.A., 2004. Meta-analysis of intellectual and
viors: the role of autonomic activity. J. Abnorm. Child Psychol. 36, 941–953. https:// neuropsychological test performance in attention-deficit/hyperactivity disorder.
doi.org/10.1007/s10802-008-9225-9. Neuropsychology 18, 543–555. https://doi.org/10.1037/0894-4105.18.3.543.
Clarke, A.R., Barry, R.J., Dupuy, F.E., McCarthy, R., Selikowitz, M., Johnstone, S.J., 2013. Geissler, J., Romanos, M., Hegerl, U., Hensch, T., 2014. Hyperactivity and sensation
Excess beta activity in the EEG of children with attention-deficit/hyperactivity dis- seeking as autoregulatory attempts to stabilize brain arousal in ADHD and mania?
order: a disorder of arousal? Int. J. Psychophysiol. 89, 314–319. https://doi.org/10. Atten. Defic. Hyperact. Disord. 6, 159–173. https://doi.org/10.1007/s12402-014-
1016/j.ijpsycho.2013.04.009. 0144-z.
Coghill, D.R., Seth, S., Pedroso, S., Usala, T., Currie, J., Gagliano, A., 2014. Effects of Gilzenrat, M.S., Nieuwenhuis, S., Jepma, M., Cohen, J.D., 2010. Pupil diameter tracks
methylphenidate on cognitive functions in children and adolescents with attention- changes in control state predicted by the adaptive gain theory of locus coeruleus
deficit/hyperactivity disorder: evidence from a systematic review and a meta-ana- function. Cogn. Affect. Behav. Neurosci. 10, 252–269. https://doi.org/10.3758/
lysis. Biol. Psychiatry 76, 603–615. https://doi.org/10.1016/j.biopsych.2013.10. CABN.10.2.252.
005. Gompf, H.S., Mathai, C., Fuller, P.M., Wood, D.A., Pedersen, N.P., Saper, C.B., Lu, J.,
Collins, P.Y., Patel, V., Joestl, S.S., March, D., Insel, T.R., Daar, A.S., Scientific Advisory, 2010. Locus ceruleus and anterior cingulate cortex sustain wakefulness in a novel
B., the Executive Committee of the Grand Challenges on Global Mental, H, Anderson, environment. J. Neurosci. 30, 14543–14551. https://doi.org/10.1523/JNEUROSCI.
W., Dhansay, M.A., Phillips, A., Shurin, S., Walport, M., Ewart, W., Savill, S.J., 3037-10.2010.
Bordin, I.A., Costello, E.J., Durkin, M., Fairburn, C., Glass, R.I., Hall, W., Huang, Y., Gray, J.A., 1981. A critique of Eysenck’s theory of personality. In: Eysenck, H.J. (Ed.), A
Hyman, S.E., Jamison, K., Kaaya, S., Kapur, S., Kleinman, A., Ogunniyi, A., Otero- Model for Personality. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 246–276.
Ojeda, A., Poo, M.M., Ravindranath, V., Sahakian, B.J., Saxena, S., Singer, P.A., Stein, https://doi.org/10.1007/978-3-642-67783-0_8.
D.J., 2011. Grand challenges in global mental health. Nature 475, 27–30. https://doi. Griffiths, K.R., Quintana, D.S., Hermens, D.F., Spooner, C., Tsang, T.W., Clarke, S., Kohn,
org/10.1038/475027a. M.R., 2017. Sustained attention and heart rate variability in children and adolescents
Conzelmann, A., Gerdes, A.B., Mucha, R.F., Weyers, P., Lesch, K.P., Bahne, C.G., with ADHD. Biol. Psychol. 124, 11–20. https://doi.org/10.1016/j.biopsycho.2017.
Fallgatter, A.J., Renner, T.J., Warnke, A., Romanos, M., Pauli, P., 2014. Autonomic 01.004.
hypoactivity in boys with attention-deficit/hyperactivity disorder and the influence Groen, Y., Mulder, L.J., Wijers, A.A., Minderaa, R.B., Althaus, M., 2009. Methylphenidate
of methylphenidate. World J. Biol. Psychiatry 15, 56–65. https://doi.org/10.3109/ improves diminished error and feedback sensitivity in ADHD: an evoked heart rate
15622975.2013.829584. analysis. Biol. Psychol. 82, 45–53. https://doi.org/10.1016/j.biopsycho.2009.05.
Critchley, H.D., Mathias, C.J., Josephs, O., O’Doherty, J., Zanini, S., Dewar, B.K., 004.
Cipolotti, L., Shallice, T., Dolan, R.J., 2003. Human cingulate cortex and autonomic Groom, M.J., Scerif, G., Liddle, P.F., Batty, M.J., Liddle, E.B., Roberts, K.L., Cahill, J.D.,
control: converging neuroimaging and clinical evidence. Brain 126, 2139–2152. Liotti, M., Hollis, C., 2010. Effects of motivation and medication on electro-
https://doi.org/10.1093/brain/awg216. physiological markers of response inhibition in children with attention-deficit/hy-
Critchley, H., Garfinkel, S., 2018. The influence of physiological signals on cognition. peractivity disorder. Biol. Psychiatry 67, 624–631. https://doi.org/10.1016/j.
Curr. Opin. Behav. Sci. 19, 13–18. https://doi.org/10.1016/j.cobeha.2017.08.014. biopsych.2009.09.029.
Crone, E.A., Jennings, J.R., van der Molen, M.W., 2003. Sensitivity to interference and Groom, M.J., Liddle, E.B., Scerif, G., Liddle, P.F., Batty, M.J., Liotti, M., Hollis, C.P., 2013.
response contingencies in attention-deficit/hyperactivity disorder. J. Child Psychol. Motivational incentives and methylphenidate enhance electrophysiological correlates
Psychiatry 44, 214–226. https://doi.org/10.1111/1469-7610.00115. of error monitoring in children with attention deficit/hyperactivity disorder. J. Child
Crowell, S.E., Beauchaine, T.P., Gatzke-Kopp, L., Sylvers, P., Mead, H., Chipman-Chacon, Psychol. Psychiatry 54, 836–845. https://doi.org/10.1111/jcpp.12069.
J., 2006. Autonomic correlates of attention-deficit/hyperactivity disorder and op- Hanc, T., Cortese, S., 2018. Attention deficit/hyperactivity-disorder and obesity: a review
positional defiant disorder in preschool children. J. Abnorm. Psychol. 115, 174–178. and model of current hypotheses explaining their comorbidity. Neurosci. Biobehav.
https://doi.org/10.1037/0021-843X.115.1.174. Rev. 92, 16–28. https://doi.org/10.1016/j.neubiorev.2018.05.017.
Cubillo, A., Halari, R., Smith, A., Taylor, E., Rubia, K., 2012. A review of fronto-striatal Hermens, D.F., Williams, L.M., Lazzaro, I., Whitmont, S., Melkonian, D., Gordon, E., 2004.
and fronto-cortical brain abnormalities in children and adults with Attention Deficit Sex differences in adult ADHD: a double dissociation in brain activity and autonomic
Hyperactivity Disorder (ADHD) and new evidence for dysfunction in adults with arousal. Biol. Psychol. 66, 221–233. https://doi.org/10.1016/j.biopsycho.2003.10.
ADHD during motivation and attention. Cortex 48, 194–215. https://doi.org/10. 006.
1016/j.cortex.2011.04.007. Hermens, D.F., Kohn, M.R., Clarke, S.D., Gordon, E., Williams, L.M., 2005a. Sex differ-
de Carvalho, T.D., Wajnsztejn, R., de Abreu, L.C., Marques Vanderlei, L.C., Godoy, M.F., ences in adolescent ADHD: findings from concurrent EEG and EDA. Clin.
Adami, F., Valenti, V.E., Monteiro, C.B.M., Leone, C., da Cruz Martins, K.C., Ferreira, Neurophysiol. 116, 1455–1463. https://doi.org/10.1016/j.clinph.2005.02.012.
C., 2014. Analysis of cardiac autonomic modulation of children with attention deficit Hermens, D.F., Williams, L.M., Clarke, S., Kohn, M., Cooper, N., Gordon, E., 2005b.
hyperactivity disorder. Neuropsychiatr. Dis. Treat. 10, 613–618. https://doi.org/10. Responses to methylphenidate in adolescent AD/HD: evidence from concurrently
2147/NDT.S49071. recorded autonomic (EDA) and central (EEG and ERP) measures. Int. J.
de Zambotti, M., Trinder, J., Silvani, A., Colrain, I.M., Baker, F.C., 2018. Dynamic cou- Psychophysiol. 58, 21–33. https://doi.org/10.1016/j.ijpsycho.2005.03.006.
pling between the central and autonomic nervous systems during sleep: A review. Herpertz, S.C., Wenning, B., Mueller, B., Qunaibi, M., Sass, H., Herpertz-Dahlmann, B.,
Neurosci. Biobehav. Rev. 90, 84–103. https://doi.org/10.1016/j.neubiorev.2018.03. 2001. Psychophysiological responses in ADHD boys with and without conduct dis-
027. order: implications for adult antisocial behavior. J. Am. Acad. Child Adolesc.
Desman, C., Petermann, F., Hampel, P., 2008. Deficit in response inhibition in children Psychiatry 40, 1222–1230. https://doi.org/10.1097/00004583-200110000-00017.
with attention deficit/hyperactivity disorder (ADHD): impact of motivation? Child Herpertz, S.C., Mueller, B., Wenning, B., Qunaibi, M., Lichterfeld, C., Herpertz-Dahlmann,
Neuropsychol. 14, 483–503. https://doi.org/10.1080/09297040701625831. B., 2003. Autonomic responses in boys with externalizing disorders. J. Neural

204
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

Transm. (Vienna) 110, 1181–1195. https://doi.org/10.1007/s00702-003-0026-6. Loo, S.K., Makeig, S., 2012. Clinical utility of EEG in attention-deficit/hyperactivity
Huang, J., Ulke, C., Sander, C., Jawinski, P., Spada, J., Hegerl, U., Hensch, T., 2018. disorder: a research update. Neurotherapeutics 9, 569–587. https://doi.org/10.1007/
Impact of brain arousal and time-on-task on autonomic nervous system activity in the s13311-012-0131-z.
wake-sleep transition. BMC Neurosci. 19, 18. https://doi.org/10.1186/s12868-018- Luman, M., Oosterlaan, J., Hyde, C., van Meel, C.S., Sergeant, J.A., 2007. Heart rate and
0419-y. reinforcement sensitivity in ADHD. J. Child Psychol. Psychiatry 48, 890–898. https://
Hvolby, A., 2015. Associations of sleep disturbance with ADHD: implications for treat- doi.org/10.1111/j.1469-7610.2007.01769.x.
ment. Atten. Defic. Hyperact. Disord. 7, 1–18. https://doi.org/10.1007/s12402-014- Luman, M., Oosterlaan, J., Sergeant, J.A., 2008. Modulation of response timing in ADHD,
0151-0. effects of reinforcement valence and magnitude. J. Abnorm. Child Psychol. 36,
Iaboni, F., Douglas, V.I., Ditto, B., 1997. Psychophysiological response of ADHD children 445–456. https://doi.org/10.1007/s10802-007-9190-8.
to reward and extinction. Psychophysiology 34, 116–123. https://doi.org/10.1111/j. Mather, M., Clewett, D., Sakaki, M., Harley, C.W., 2016. Norepinephrine ignites local
1469-8986.1997.tb02422.x. hotspots of neuronal excitation: how arousal amplifies selectivity in perception and
Isaksson, J., Nilsson, K.W., Nyberg, F., Hogmark, A., Lindblad, F., 2012. Cortisol levels in memory. Behav. Brain Sci. 39, e200. https://doi.org/10.1017/S0140525X15000667.
children with attention-deficit/hyperactivity disorder. J. Psychiatr. Res. 46, Matthews, S.C., Paulus, M.P., Simmons, A.N., Nelesen, R.A., Dimsdale, J.E., 2004.
1398–1405. https://doi.org/10.1016/j.jpsychires.2012.08.021. Functional subdivisions within anterior cingulate cortex and their relationship to
James, S.N., Cheung, C.H.M., Rijsdijk, F., Asherson, P., Kuntsi, J., 2016. Modifiable autonomic nervous system function. NeuroImage 22, 1151–1156. https://doi.org/10.
arousal in Attention-Deficit/Hyperactivity disorder and its etiological association 1016/j.neuroimage.2004.03.005.
with fluctuating reaction times. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 1, Mayer, K., Wyckoff, S.N., Strehl, U., 2016. Underarousal in adult ADHD: how are per-
539–547. https://doi.org/10.1016/j.bpsc.2016.06.003. ipheral and cortical arousal related? Clin. EEG Neurosci. 47, 171–179. https://doi.
Jennings, J.R., vanderMolen, M.W., Pelham, W., Debski, K.B., Hoza, B., 1997. Inhibition org/10.1177/1550059415577544.
in boys with attention deficit hyperactivity disorder as indexed by heart rate change. McQuade, J.D., Breaux, R.P., 2017. Are elevations in ADHD symptoms associated with
Dev. Psychol. 33, 308–318. https://doi.org/10.1037/0012-1649.33.2.308. physiological reactivity and emotion dysregulation in children? J. Abnorm. Child
Jensen, C.M., Steinhausen, H.C., 2015. Comorbid mental disorders in children and ado- Psychol. 45, 1091–1103. https://doi.org/10.1007/s10802-016-0227-8.
lescents with attention-deficit/hyperactivity disorder in a large nationwide study. McQuade, J.D., Penzel, T.E., Silk, J.S., Lee, K.H., 2017. Parasympathetic nervous system
Atten. Defic. Hyperact. Disord. 7, 27–38. https://doi.org/10.1007/s12402-014- reactivity moderates associations between children’s executive functioning and social
0142-1. and academic competence. J. Abnorm. Child Psychol. 45, 1355–1367. https://doi.
Johnson, K.A., Wiersema, J.R., Kuntsi, J., 2009. What would Karl Popper say? Are current org/10.1007/s10802-016-0246-5.
psychological theories of ADHD falsifiable? Behav. Brain Funct. 5, 15. https://doi. Metin, B., Roeyers, H., Wiersema, J.R., van der Meere, J., Sonuga-Barke, E., 2012. A meta-
org/10.1186/1744-9081-5-15. analytic study of event rate effects on Go/No-Go performance in attention-deficit/
Johnstone, S.J., Watt, A.J., Dimoska, A., 2010. Varying required effort during interference hyperactivity disorder. Biol. Psychiatry 72, 990–996. https://doi.org/10.1016/j.
control in children with AD/HD: task performance and ERPs. Int. J. Psychophysiol. biopsych.2012.08.023.
76, 174–185. https://doi.org/10.1016/j.ijpsycho.2010.03.010. Minzenberg, M.J., Watrous, A.J., Yoon, J.H., Ursu, S., Carter, C.S., 2008. Modafinil shifts
Johnstone, S.J., Barry, R.J., Clarke, A.R., 2013. Ten years on: a follow-up review of ERP human locus coeruleus to low-tonic, high-phasic activity during functional MRI.
research in attention-deficit/hyperactivity disorder. Clin. Neurophysiol. 124, Science (New York, N.Y.) 322, 1700–1702. https://doi.org/10.1126/science.
644–657. https://doi.org/10.1016/j.clinph.2012.09.006. 1164908.
Kara, K., Karaman, D., Erdem, U., Congologlu, M.A., Durukan, I., Ilhan, A., 2013. Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., Group, P., 2009. Preferred reporting
Investigation of autonomic nervous system functions by pupillometry in children items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 6,
with Attention-Deficit/Hyperactivity disorder. Klin Psikofarmakol B 23, 49–56. e1000097. https://doi.org/10.1371/journal.pmed.1000097.
https://doi.org/10.5455/bcp.20121130085850. Mueller, A., Hong, D.S., Shepard, S., Moore, T., 2017. Linking ADHD to the neural cir-
Karalunas, S.L., Geurts, H.M., Konrad, K., Bender, S., Nigg, J.T., 2014. Annual research cuitry of attention. Trends Cogn. Sci. 21, 474–488. https://doi.org/10.1016/j.tics.
review: reaction time variability in ADHD and autism spectrum disorders: measure- 2017.03.009.
ment and mechanisms of a proposed trans-diagnostic phenotype. J. Child Psychol. Mukhopadhyay, P., De, S., Roychowdhuri, A., Chakraborti, M., Chattopadhyay, P., 1997.
Psychiatry 55, 685–710. https://doi.org/10.1111/jcpp.12217. Arousal modulation capacity in the deficit disorders. Soc. Sci. Int. 13, 25–30.
Karatekin, C., Bingham, C., White, T., 2010. Oculomotor and pupillometric indices of pro- Murphy, P.R., O’Connell, R.G., O’Sullivan, M., Robertson, I.H., Balsters, J.H., 2014. Pupil
and antisaccade performance in youth-onset psychosis and attention deficit/hyper- diameter covaries with BOLD activity in human locus coeruleus. Hum. Brain Mapp.
activity disorder. Schizophr. Bull. 36, 1167–1186. https://doi.org/10.1093/schbul/ 35, 4140–4154. https://doi.org/10.1002/hbm.22466.
sbp035. Murray, E.A., 2007. The amygdala, reward and emotion. Trends Cogn. Sci. 11, 489–497.
Keage, H.A., Clark, C.R., Hermens, D.F., Kohn, M.R., Clarke, S., Williams, L.M., Crewther, https://doi.org/10.1016/j.tics.2007.08.013.
D., Lamb, C., Gordon, E., 2006. Distractibility in AD/HD predominantly inattentive Musser, E.D., Backs, R.W., Schmitt, C.F., Ablow, J.C., Measelle, J.R., Nigg, J.T., 2011.
and combined subtypes: the P3a ERP component, heart rate and performance. J. Emotion regulation via the autonomic nervous system in children with attention-
Integr. Neurosci. 5, 139–158. https://doi.org/10.1142/S0219635206001070. deficit/hyperactivity disorder (ADHD). J. Abnorm. Child Psychol. 39, 841–852.
Kofler, M.J., Rapport, M.D., Sarver, D.E., Raiker, J.S., Orban, S.A., Friedman, L.M., https://doi.org/10.1007/s10802-011-9499-1.
Kolomeyer, E.G., 2013. Reaction time variability in ADHD: a meta-analytic review of Musser, E.D., Galloway-Long, H.S., Frick, P.J., Nigg, J.T., 2013. Emotion regulation and
319 studies. Clin. Psychol. Rev. 33, 795–811. https://doi.org/10.1016/j.cpr.2013.06. heterogeneity in attention-deficit/hyperactivity disorder. J. Am. Acad. Child Adolesc.
001. Psychiatry 52 (163-171), e162. https://doi.org/10.1016/j.jaac.2012.11.009.
Kuntsi, J., Klein, C., 2012. Intraindividual variability in ADHD and its implications for Musser, E.D., Lugo, Y., Ward, A.R., Tenenbaum, R.B., Morris, S., Brijmohan, N., Martinez,
research of causal links. Curr. Top. Behav. Neurosci. 9, 67–91. https://doi.org/10. J., 2018. Parent emotion expression and autonomic-linked emotion dysregulation in
1007/7854_2011_145. childhood ADHD. J. Psychopathol. Behav. Assess. 40, 593–605. https://doi.org/10.
Lackschewitz, H., Huther, G., Kroner-Herwig, B., 2008. Physiological and psychological 1007/s10862-018-9685-3.
stress responses in adults with attention-deficit/hyperactivity disorder (ADHD). Nagai, Y., Goldstein, L.H., Critchley, H.D., Fenwick, P.B., 2004. Influence of sympathetic
Psychoneuroendocrinology 33, 612–624. https://doi.org/10.1016/j.psyneuen.2008. autonomic arousal on cortical arousal: implications for a therapeutic behavioural
01.016. intervention in epilepsy. Epilepsy Res. 58, 185–193. https://doi.org/10.1016/j.
Lawrence, C.A., Barry, R.J., Clarke, A.R., Johnstone, S.J., McCarthy, R., Selikowitz, M., eplepsyres.2004.02.004.
Broyd, S.J., 2005. Methylphenidate effects in attention deficit/hyperactivity disorder: National Institute for Health and Care Excellence, 2018. Attention Deficit Hyperactivity
electrodermal and ERP measures during a continuous performance task. Disorder: Diagnosis and Management. https://www.nice.org.uk/guidance/ng87.
Psychopharmacology (Berl.) 183, 81–91. https://doi.org/10.1007/s00213-005- Newson, J.J., Thiagarajan, T.C., 2018. EEG frequency bands in psychiatric disorders: a
0144-y. review of resting state studies. Front. Hum. Neurosci. 12, 521. https://doi.org/10.
Lazzaro, I., Gordon, E., Li, W., Lim, C.L., Plahn, M., Whitmont, S., Clarke, S., Barry, R.J., 3389/fnhum.2018.00521.
Dosen, A., Meares, R., 1999. Simultaneous EEG and EDA measures in adolescent Ng, Q.X., Ho, C.Y.X., Chan, H.W., Yong, B.Z.J., Yeo, W.S., 2017. Managing childhood and
attention deficit hyperactivity disorder. Int. J. Psychophysiol. 34, 123–134. https:// adolescent attention-deficit/hyperactivity disorder (ADHD) with exercise: a sys-
doi.org/10.1016/S0167-8760(99)00068-9. tematic review. Complement. Ther. Med. 34, 123–128. https://doi.org/10.1016/j.
Leikauf, J.E., Griffiths, K.R., Saggar, M., Hong, D.S., Clarke, S., Efron, D., Tsang, T.W., ctim.2017.08.018.
Hermens, D.F., Kohn, M.R., Williams, L.M., 2017. Identification of biotypes in at- O’Connell, R.G., Bellgrove, M.A., Dockree, P.M., Robertson, I.H., 2004. Reduced elec-
tention-deficit/hyperactivity disorder, a report from a randomized, controlled trial. trodermal response to errors predicts poor sustained attention performance in at-
Pers. Med. Psychiatry 3, 8–17. https://doi.org/10.1016/j.pmip.2017.02.001. tention deficit hyperactivity disorder. Neuroreport 15, 2535–2538. https://doi.org/
Liang, E.F., Lim, S.Z., Tam, W.W., Ho, C.S., Zhang, M.W., McIntyre, R.S., Ho, R.C., 2018. 10.1097/00001756-200411150-00021.
The effect of methylphenidate and Atomoxetine on heart rate and systolic blood Oliver, M.L., Nigg, J.T., Cassavaugh, N.D., Backs, R.W., 2012. Behavioral and cardio-
pressure in young people and adults with attention-deficit hyperactivity disorder vascular responses to frustration during simulated driving tasks in young adults with
(ADHD): systematic review, meta-analysis, and meta-regression. Int. J. Environ. Res. and without attention disorder symptoms. J. Atten. Disord. 16, 478–490. https://doi.
Public Health 15. https://doi.org/10.3390/ijerph15081789. org/10.1177/1087054710397132.
Liddle, E.B., Hollis, C., Batty, M.J., Groom, M.J., Totman, J.J., Liotti, M., Scerif, G., Liddle, Perrin, P.B., Case, K.H., Byrd, D.L., Snipes, D.J., Anderson, K.L., Berg, W.K., 2014.
P.F., 2011. Task-related default mode network modulation and inhibitory control in Executive functioning in attention-deficit/hyperactivity disorder: questioning the
ADHD: effects of motivation and methylphenidate. J. Child Psychol. Psychiatry 52, notion of planning deficits with heart rate reactivity. Atten. Defic. Hyperact. Disord.
761–771. https://doi.org/10.1111/j.1469-7610.2010.02333.x. 6, 1–10. https://doi.org/10.1007/s12402-013-0118-6.
Liotti, M., Pliszka, S.R., Perez, R., Kothmann, D., Woldorff, M.G., 2005. Abnormal brain Polanczyk, G.V., Willcutt, E.G., Salum, G.A., Kieling, C., Rohde, L.A., 2014. ADHD pre-
activity related to performance monitoring and error detection in children with valence estimates across three decades: an updated systematic review and meta-re-
ADHD. Cortex 41, 377–388. https://doi.org/10.1016/S0010-9452(08)70274-0. gression analysis. Int. J. Epidemiol. 43, 434–442. https://doi.org/10.1093/ije/

205
A. Bellato, et al. Neuroscience and Biobehavioral Reviews 108 (2020) 182–206

dyt261. 2015. Methylphenidate for attention-deficit/hyperactivity disorder in children and

Porges, S.W., 2009. The polyvagal theory: new insights into adaptive reactions of the adolescents: cochrane systematic review with meta-analyses and trial sequential
autonomic nervous system. Cleve. Clin. J. Med. 76 (Suppl 2), S86–90. https://doi. analyses of randomised clinical trials. BMJ (Clin. Res. ed.) 351, h5203. https://doi.
org/10.3949/ccjm.76.s2.17. org/10.1136/bmj.h5203.
Rajkowski, J., 1993. Correlations Between Locus Coeruleus (LC) Neural Activity, Pupil Strauß, M., Ulke, C., Paucke, M., Huang, J., Mauche, N., Sander, C., Stark, T., Hegerl, U.,
Diameter and Behavior in Monkey Support a Role of LC in Attention. 1993. Soc. 2018. Brain arousal regulation in adults with attention-deficit/hyperactivity disorder
Neurosc., Abstract, Washington, D. C. https://ci.nii.ac.jp/naid/10021384962/en/. (ADHD). Psychiatry Res. 261, 102–108. https://doi.org/10.1016/j.psychres.2017.12.
Rubia, K., 2018. Cognitive neuroscience of attention deficit hyperactivity disorder 043.
(ADHD) and its clinical translation. Front. Hum. Neurosci. 12, 100. https://doi.org/ Tenenbaum, R.B., Musser, E.D., Raiker, J.S., Coles, E.K., Gnagy, E.M., Pelham, W.E.,
10.3389/fnhum.2018.00100. 2018. Specificity of reward sensitivity and parasympathetic-based regulation among
Rubia, K., Criaud, M., Wulff, M., Alegria, A., Brinson, H., Barker, G., Stahl, D., Giampietro, children with Attention-Deficit/Hyperactivity and disruptive behavior disorders. J.
V., 2019. Functional connectivity changes associated with fMRI neurofeedback of Abnorm. Child Psychol. 46, 965–977. https://doi.org/10.1007/s10802-017-0343-0.
right inferior frontal cortex in adolescents with ADHD. NeuroImage 188, 43–58. Van Der Meere, J., 2002. The role of attention. In: Sandberg, S. (Ed.), Monographs on
https://doi.org/10.1016/j.neuroimage.2018.11.055. Child and Adolescent Psychiatry: Hyperactivity Disorders, 2nd ed. Cambridge
Rukmani, M.R., Seshadri, S.P., Thennarasu, K., Raju, T.R., Sathyaprabha, T.N., 2016. University Press, pp. 162–213. http://hdl.handle.net/11370/2a95c0c0-0cf7-4046-
Heart rate variability in children with Attention-Deficit/Hyperactivity disorder: a b4d0-3d0faa9a7d06.
pilot study. Ann. Neurosci. 23, 81–88. https://doi.org/10.1159/000443574. Van Der Meere, J., Börger, N., Wiersema, J., 2010. ADHD: state regulation and motiva-
Saad, J.F., Kohn, M.R., Clarke, S., Lagopoulos, J., Hermens, D.F., 2015. Is the Theta/Beta tion. CML Psychiatry 21, 14–20. http://hdl.handle.net/11370/9a02cefb-c792-4f12-
EEG Marker for ADHD Inherently Flawed? J. Atten. Disord. 22, 815–826. https://doi. bc08-acd0189859ac.
org/10.1177/1087054715578270. Van Dessel, J., Sonuga-Barke, E., Mies, G., Lemiere, J., Van der Oord, S., Morsink, S.,
Samuels, E.R., Szabadi, E., 2008. Functional neuroanatomy of the noradrenergic locus Danckaerts, M., 2018. Delay aversion in attention deficit/hyperactivity disorder is
coeruleus: its roles in the regulation of arousal and autonomic function part II: mediated by amygdala and prefrontal cortex hyper-activation. J. Child Psychol.
physiological and pharmacological manipulations and pathological alterations of Psychiatry 59, 888–899. https://doi.org/10.1111/jcpp.12868.
locus coeruleus activity in humans. Curr. Neuropharmacol. 6, 254–285. https://doi. Wainstein, G., Rojas-Libano, D., Crossley, N.A., Carrasco, X., Aboitiz, F., Ossandon, T.,
org/10.2174/157015908785777193. 2017. Pupil size tracks attentional performance in Attention-Deficit/Hyperactivity
Sanders, A.F., 1983. Towards a model of stress and human performance. Acta Psychol. 53, disorder. Sci. Rep. 7, 8228. https://doi.org/10.1038/s41598-017-08246-w.
61–97. Wang, T.S., Huang, W.L., Kuo, T.B.J., Lee, G.S., Yang, C.C.H., 2013. Inattentive and hy-
Sara, S.J., Bouret, S., 2012. Orienting and reorienting: the locus coeruleus mediates peractive preschool-age boys have lower sympathetic and higher parasympathetic
cognition through arousal. Neuron 76, 130–141. https://doi.org/10.1016/j.neuron. activity. J. Physiol. Sci. 63, 87–94. https://doi.org/10.1007/s12576-012-0238-3.
2012.09.011. Wang, X., Piñol, R.A., Byrne, P., Mendelowitz, D., 2014. Optogenetic stimulation of locus
Satterfield, J.H., Schell, A.M., Backs, R.W., Hidaka, K.C., 1984. A cross-sectional and ceruleus neurons augments inhibitory transmission to parasympathetic cardiac vagal
longitudinal study of age effects of electrophysiological measures in hyperactive and neurons via activation of brainstem α1 and β1 receptors. J. Neurosci. 34, 6182–6189.
normal children. Biol. Psychiatry 19, 973–990. https://www.ncbi.nlm.nih.gov/ https://doi.org/10.1523/jneurosci.5093-13.2014.
pubmed/6332646. Ward, A.R., Alarcon, G., Nigg, J.T., Musser, E.D., 2015. Variation in parasympathetic
Sayer, G.R., McGough, J.J., Levitt, J., Cowen, J., Sturm, A., Castelo, E., McCracken, J.T., dysregulation moderates short-term memory problems in childhood Attention-
2016. Acute and long-term cardiovascular effects of stimulant, guanfacine, and Deficit/Hyperactivity disorder. J. Abnorm. Child Psychol. 43, 1573–1583. https://
combination therapy for attention-deficit/hyperactivity disorder. J. Child Adolesc. doi.org/10.1007/s10802-015-0054-3.
Psychopharmacol. 26, 882–888. https://doi.org/10.1089/cap.2015.0264. Waschbusch, D.A., Pelham Jr., W.E., Jennings, J.R., Greiner, A.R., Tarter, R.E., Moss,
Sellaro, R., van Leusden, J.W., Tona, K.D., Verkuil, B., Nieuwenhuis, S., Colzato, L.S., H.B., 2002. Reactive aggression in boys with disruptive behavior disorders: behavior,
2015. Transcutaneous vagus nerve stimulation enhances post-error slowing. J. Cogn. physiology, and affect. J. Abnorm. Child Psychol. 30, 641–656. https://doi.org/10.
Neurosci. 27, 2126–2132. https://doi.org/10.1162/jocn_a_00851. 1023/A:1020867831811.
Sergeant, J., 2000. The cognitive-energetic model: an empirical approach to attention- Wass, S.V., de Barbaro, K., Clackson, K., 2015. Tonic and phasic co-variation of peripheral
deficit hyperactivity disorder. Neurosci. Biobehav. Rev. 24, 7–12. https://doi.org/10. arousal indices in infants. Biol. Psychol. 111, 26–39. https://doi.org/10.1016/j.
1016/S0149-7634(99)00060-3. biopsycho.2015.08.006.
Sergeant, J.A., 2005. Modeling attention-deficit/hyperactivity disorder: a critical ap- Wiersema, R., van der Meere, J., Roeyers, H., Van Coster, R., Baeyens, D., 2006. Event
praisal of the cognitive-energetic model. Biol. Psychiatry 57, 1248–1255. https://doi. rate and event-related potentials in ADHD. J. Child Psychol. Psychiatry 47, 560–567.
org/10.1016/j.biopsych.2004.09.010. https://doi.org/10.1111/j.1469-7610.2005.01592.x.
Shibagaki, M., Yamanaka, T., Furuya, T., 1993. Attention state in electrodermal activity Wilbertz, G., Trueg, A., Sonuga-Barke, E.J., Blechert, J., Philipsen, A., Tebartz van Elst, L.,
during auditory stimulation of children with attention-deficit hyperactivity disorder. 2013. Neural and psychophysiological markers of delay aversion in attention-deficit
Percept. Mot. Skills 77, 331–338. https://doi.org/10.2466/pms.1993.77.1.331. hyperactivity disorder. J. Abnorm. Psychol. 122, 566–572. https://doi.org/10.1037/
Shibagaki, M., Furuya, T., 1997. Baseline respiratory sinus arrhythmia and heart-rate a0031924.
responses during auditory stimulation of children with attention-deficit hyperactivity Wilbertz, G., Delgado, M.R., Tebartz Van Elst, L., Maier, S., Philipsen, A., Blechert, J.,
disorder. Percept. Mot. Skills 84, 967–975. https://doi.org/10.2466/pms.1997.84.3. 2017. Neural response during anticipation of monetary loss is elevated in adult at-
967. tention deficit hyperactivity disorder. World J. Biol. Psychiatry 18, 268–278. https://
Shiels, K., Hawk Jr, L.W., 2010. Self-regulation in ADHD: the role of error processing. doi.org/10.3109/15622975.2015.1112032.
Clin. Psychol. Rev. 30, 951–961. https://doi.org/10.1016/j.cpr.2010.06.010. Willcutt, E.G., Doyle, A.E., Nigg, J.T., Faraone, S.V., Pennington, B.F., 2005. Validity of
Sonuga-Barke, E.J., 2002. Psychological heterogeneity in AD/HD - a dual pathway model the executive function theory of attention-deficit/hyperactivity disorder: a meta-
of behaviour and cognition. Behav. Brain Res. 130, 29–36. https://doi.org/10.1016/ analytic review. Biol. Psychiatry 57, 1336–1346. https://doi.org/10.1016/j.
S0166-4328(01)00432-6. biopsych.2005.02.006.
Sonuga-Barke, E.J., Halperin, J.M., 2010. Developmental phenotypes and causal path- Yerkes, R.M., Dodson, J.D., 1908. The relation of strength of stimulus to rapidity of habit-
ways in attention deficit/hyperactivity disorder: potential targets for early inter- formation. J. Comp. Neurol. Psychol. 18, 459–482. https://doi.org/10.1002/cne.
vention? J. Child Psychol. Psychiatry 51, 368–389. https://doi.org/10.1111/j.1469- 920180503.
7610.2009.02195.x. Zhang, S., Hu, S., Chao, H.H., Ide, J.S., Luo, X., Farr, O.M., Li, C.S., 2014. Ventromedial
Storebo, O.J., Krogh, H.B., Ramstad, E., Moreira-Maia, C.R., Holmskov, M., Skoog, M., prefrontal cortex and the regulation of physiological arousal. Soc. Cogn. Affect.
Nilausen, T.D., Magnusson, F.L., Zwi, M., Gillies, D., Rosendal, S., Groth, C., Neurosci. 9, 900–908. https://doi.org/10.1093/scan/nst064.
Rasmussen, K.B., Gauci, D., Kirubakaran, R., Forsbol, B., Simonsen, E., Gluud, C.,

206