Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent disorders in childhood and adolescence (Polanczyk et al., 2007). The clinical picture is characterized by age-inappropriate levels of impulsivity, motor activity, and deficient attentional capacity. The latter deficit manifests in severe difficulties to direct and maintain attention as well as a significantly enhanced distractibility.

Distractibility is however not exclusively associated with negative consequences but is an important mechanism to promote adaptive behavior. Thus, the potential positive impact of unexpected stimuli has been attributed to the contextual novelty of these distractors (e.g., SanMiguel et al., 2010; Wetzel et al., 2012; Schomaker and Meeter, 2014, 2015). Novel stimuli are highly salient as their appearance might indicate the necessity of behavioral adaptation to avoid harm or gain a potential reward. Therefore, novel as well as unexpected (contextually novel) stimuli elicit an automatic orienting response (Sokolov, 1963) and simultaneously increase alertness (Aston-Jones and Cohen, 2005). If the alerting benefits of the disruption exceed the orienting costs, behavioral facilitation can occur, otherwise performance detriments should be observable (SanMiguel et al., 2010).

This conceptual framework could be of significance for understanding different dysfunctions in attentional processes in ADHD. On the one hand, deficient inhibition of external stimulation has been thought to be responsible for the breakthrough of task-irrelevant information (Barkley, 1997), suggesting enhanced orienting costs (distractibility) in ADHD patients. On the other hand, openness to external stimulation has been viewed as a mean to overcome underlying deficits of arousal and alertness in ADHD (Zentall and Zentall, 1983; Sergeant, 2005). In this regard, alerting benefits, e.g., utilization of warning cues, can be of utmost importance.

Previous studies in ADHD patients revealed the divergent effects of task-unrelated, especially novel stimuli on behavior. Some studies reported the anticipated distracting effect of task-irrelevant stimuli on children, adolescents as well as adults with ADHD (e.g., Gumenyuk et al., 2005; Berger and Cassuto, 2014; Pelletier et al., 2016). Others showed, however, that a novel environment reduced hyperactive behavior in waiting situations (Antrop et al., 2000) and that novel sounds facilitated attentional processing in visual decision tasks in ADHD patients (van Mourik et al., 2007; Tegelbeckers et al., 2016). In these latter studies, novel sounds preceded the onset of a task-display so that the alerting effect of novelty could unfold optimally and reduce omission errors (van Mourik et al., 2007) as well as commission errors and reaction time variability (Tegelbeckers et al., 2016).

Most importantly, the beneficial effects of novel sounds seem to exceed the effect of a repeatedly presented, familiar sound (standard) on behavioral attentional performance (van Mourik et al., 2007; Tegelbeckers et al., 2016). Facilitation in this case can thus not purely be attributed to the preparatory signal provided by any sound cue. Instead, the unexpectedness of the unique sound seems to play a crucial role causing either enhanced alerting benefits or the orienting reaction, which is only triggered by novel but not familiar stimuli.

Here, we aimed for the first time to gain insight into the neural processes underlying the beneficial effects of task-preceding novel sounds in children and adolescents with but also without ADHD. Following SanMiguel et al. (2010), we expected that novel sounds operate via different attention related processes based on separate and distinguishable neural networks (Petersen and Posner, 2012).

The fronto-parietal alerting network is closely related to noradrenergic brain circuits and refers to the ability to maintain an alert state for a certain period of time (tonic) as well as to utilize external cues to increase the readiness to respond (phasic). The behavioral advantage of using phasic alerting cues seems to be unimpaired in children with ADHD as they benefit from task announcing information similar to typically developing (TD) children (e.g., Gupta and Kar, 2009; Casagrande et al., 2012; Fassbender et al., 2015). The level of tonic alertness is, on the other hand, supposedly decreased in ADHD (Zentall and Zentall, 1983; Sergeant, 2000) resulting in diminished sustained attention over time. Thus, the positive effect of task-preceding sounds in ADHD patients could be due to phasic activation increases of the tonic network.

Furthermore, two orienting networks are responsible for the voluntary direction of attention toward a task at hand (top-down) as well as the automatic allocation of attention toward changes in the environment (bottom-up) (Corbetta et al., 2008). These two functionally distinguishable components are subserved by different brain networks: bottom-up processing and automatic allocation of attention to salient sensory stimuli involves the temporo-parietal junction (TPJ) and ventral frontal cortex (ventral attention network, VAN) whereas top-down control of attention requires a dorsally located network (dorsal attention network, DAN) consisting of the intraparietal sulcus (IPS) and frontal eye field (FEF) (Corbetta and Shulman, 2002). Previous studies in ADHD revealed impairments during target processing which indicates deficits in controlled top-down orienting of attention (Kemner et al., 1996; Jonkman et al., 2000). The evidence for an ADHD related deficit in automatic orienting, on the contrary, remains scarce and inconsistent (Barry et al., 2003; Huang-Pollock and Nigg, 2003; Cortese et al., 2012). We previously found evidence for both, missing suppression of VAN activity as well as diminished habituation of novelty processing in ADHD patients when we disentangled effects of task relevance, rareness and novelty (Tegelbeckers et al., 2015). These results point to an elevated effect of novelty which might as well be associated with the beneficial behavioral effects of novel sounds in ADHD patients.

Here, we were interested in how alerting and/or orienting responses caused by novel sounds would influence brain activity and behavior during the subsequent execution of a visual flanker task (Tegelbeckers et al., 2016) in a group of typically developing children and adolescents as well as in a group of patients with ADHD who suffer from increased distractibility. To disentangle novelty related effects on task performance from mere alertness by a warning cue and to investigate the role of the proposed alerting deficit in ADHD (Sergeant, 2005), we compared the general alerting effect of a sound cue (standard vs. no sound) as well as the influence of novel sounds to the baseline without sound and extracted the specific novelty effect (novel vs. standard sound).

We expected that all participants benefit from sound presentations but that attentional performance is increased to a larger extent in ADHD patients for both warning sounds due to the alerting deficit associated with the disorder (Sergeant, 2005) and their enhanced susceptibility to novelty (Tegelbeckers et al., 2015, 2016). Thus, we also expected increased activity in the fronto-parietal alerting network in the patient group. Moreover, we hypothesized that novel sounds elicit activity in the fronto-temporoparietal (ventral) attention network in all participants but more strongly in ADHD patients. We aimed to identify the beneficial relationships underlying these novelty related activations and task performance.

The local ethics committee of the university, faculty of medicine, approved the current study and confirmed its accordance with the ethical standards of the Declaration of Helsinki. All children and adolescents as well as their parents gave written assent/consent. The participants were reimbursed with 5€ - gift vouchers per hour for a local shopping center.

The current study aimed to identify activity in neural attention networks that underlie the beneficial effects of task preceding sounds on subsequent performance in children and adolescents with and without ADHD. We were particularly interested in the effects of unique (novel) sounds compared to a repeatedly presented standard as both types of sounds should function as alerting cues, but only novel sounds were supposed to elicit an orienting response.

We found that novel sounds caused activity in large bilateral clusters spanning the STGs extending to anterior TPJs as well as the right IFG in all participants, without detectable alterations in patients with ADHD. This network is in line with former descriptions of the ventral attention (orienting) network (VAN) (Corbetta and Shulman, 2002; Corbetta et al., 2008; Petersen and Posner, 2012) but also covers the bilateral auditory cortex. In contrast to most previous reports, activation in the fronto-temporoparietal VAN was not paralleled by disruption of task performance in either group. Instead, novel sounds had a beneficial effect on reaction times and accuracy compared to the baseline without sound. Moreover, the difference between rIFG neural activity for novel minus standard sounds was correlated with the benefit in speed for novel vs. standard sound trials during the flanker task. This relationship indicates a direct facilitating effect of brain activity on subsequent task behavior caused by novelty.

Importantly, although the increase in accuracy for novel sounds in patients with ADHD exceeded the effect of novelty in TD peers, the ADHD group did not display a stronger neural orienting response than the comparison group. This neural activation overlap is in contrast to reports of enhanced sensitivity for novelty in ADHD patients in oddball tasks (Gumenyuk et al., 2005; Tegelbeckers et al., 2015). In these tasks, novelty processing is part of the task and thus influenced by potential motor reactions and target identification. Ongoing executive processes such as response selection, execution, and/or inhibition make it therefore difficult to disentangle attentional orienting effects of novelty from task related processes. Top-down attentional processes should be significantly more involved in studies using oddball designs compared to the current setup and could be the source of ADHD related impairments (Tegelbeckers et al., 2015) rather than enhanced bottom-up reactions of the fronto-temporo-parietal VAN. Moreover, the occurrence probability of a novel stimulus is significantly reduced in oddball tasks compared to the current study in which we presented novel sounds with the same probability as standard sounds. As the influence of novel stimuli possibly depends on the degree of their deviation from context (Schomaker and Meeter, 2015), group differences might become more prominent with lower frequency of novel events (Tegelbeckers et al., 2015). It is therefore likely that the context in which novel sounds occur (or orienting in general is provoked) plays an important role for alterations in ADHD. Moreover, our current finding of intact neural orienting following novel sounds in ADHD is in line with other previous reports of efficient utilization of orienting cues in the visuo-spatial domain in patients with ADHD (e.g., Huang-Pollock and Nigg, 2003; Oberlin et al., 2005; Booth et al., 2007; Mullane et al., 2011). Evidence accumulates that these patients are not impaired in filtering irrelevant sounds when engaged in a task (Friedman-Hill, 2010) and are even able to use these sounds to improve performance (van Mourik et al., 2007; Tegelbeckers et al., 2016).

We thus combined the groups for a total of 50 participants to gain a new perspective on the potential general neural mechanisms underlying these task improvements: When we and others showed behavioral facilitation by novel sounds before (van Mourik et al., 2007; Tegelbeckers et al., 2016), we argued that the alerting benefits associated with novelty might cause these improvements. Benefits in terms of lower reaction times or higher accuracy that are associated with activity in the fronto-temporopartietal VAN seem to contradict the assumed model of “orienting costs” and “alerting benefits” (SanMiguel et al., 2010). Still the novelty related activations in the bilateral STG and rIFG overlap largely with the previously described “orienting network” (Corbetta et al., 2008; Petersen and Posner, 2012).

Here, we propose that at least three explanations can account for beneficial effects of orienting network activity during task performance. First, activation associated with attentional orienting might carry an internal alerting effect in itself that is independent of the alerting network but might also be modulated by noradrenergic pathways. Corbetta et al. (2008) suggested a functional connection between noradrenergic locus coeruleus neurons and the fronto-temporoparietal VAN because both systems show similar activation pattern during deviance detection. Alternatively, the distracting effect of attentional orienting might be time dependent. Our delay of 600 ms between sound onset and target appearance might have enabled the redirection of attention without any behavioral cost. An orienting response concurrent with target presentation might have a different (more distracting) effect (Berger and Cassuto, 2014). Finally, the role of the fronto-temporoparietal VAN might as well not be limited to attentional orienting. This presumption is supported by investigations showing that VAN activity following sensory stimuli is driven by task-relevance rather than stimulus salience (Corbetta et al., 2008). As these findings contradict pure bottom-up transfer, it has been speculated that these signals indicate transitions between two tasks or behaviors (for review see Corbetta et al., 2008 or Vossel et al., 2014). Thus, as novel sounds in the current experiment serve as temporal cues announcing the task, fronto-temporoparietal VAN activity during the flanker task might reflect contextual updating of the switch between auditory and visual processing. The effects of novel sounds might then indeed rely on VAN activity characterized as “orienting response” in so far as the current task set is re-installed and behavioral execution is facilitated.

Interestingly, within the network the rIFG seems to play a crucial role of directly controlling behavior as rIFG activity was correlated with the benefit in reaction times. This fits the proposed role of the rIFG as a hub for executive control, particularly for successful attentional control and inhibition (Hampshire et al., 2010). It seems likely that the increase in attentional control via activation of the rIFG which is caused by orienting toward novelty is responsible for the benefits in task performance we observe in this setup. Moreover, structural (Sowell et al., 2003; Durston et al., 2004) and functional (Aron and Poldrack, 2005; Rubia et al., 2011) alterations of the rIFG are commonly reported in patients with ADHD, indicating a particular role of this region in the attentional problems associated with the disorder. Our findings of a modulation of task performance by an increase in rIFG activity further supports the choice of this area as a target for non-invasive brain stimulation (Breitling et al., 2016).

Finally, in addition to the novelty related findings, we also investigated the alerting effects of the frequently presented standard sound. Standard sounds improved accuracy and accelerated the responses to a similar extent as novel sounds. The ADHD group seemed to particularly benefit in accuracy also by the standard which illustrates once more that the disorder is characterized by an underlying alerting deficit (Sergeant, 2000). The lower general level of alertness probably enabled ADHD patients to improve to a greater extent than TD peers in trials which contained a warning stimulus. Moreover, the greater behavioral alerting benefits in the ADHD compared to TD group were paralleled by enhanced neural activity for standard sound trials vs. the baseline without sound in the left insula extending to putamen. As the insula is known to be involved in the integration of salient information and coordination of attentional resources (Menon and Uddin, 2010), insular dysfunction during cognitive tasks has been associated with ADHD before (Cortese et al., 2012). Here, we showed that repeatedly presented sound cues could increase insular activation and potentially normalize behavioral performance. Moreover, both groups showed activity pattern in the bilateral STG and auditory cortices for trials with preceding standard cue which provides further evidence for the idea that “alerting network” activity depends on modality specific brain areas that rely on the respective primary sensory cortices and processing pathways of the cue modality (De Santis et al., 2007; Langner et al., 2012; Spagna et al., 2015).

Taken together, the current experiment investigated brain responses associated with the beneficial effects of task-preceding standard and novel sounds in children and adolescents with ADHD and a typically developing comparison group. Although carefully designed, the study in its current form still faces some limitations. First, it is possible that we failed to find novelty related differences between ADHD and TD participants due to the sample size (type II error) or specific characteristics of the ADHD sample. We only included males and about half of our ADHD sample had used stimulant medication over the lifespan. Twelve patients currently took medication and although they discontinued the intake for at least 24 h, it cannot be ruled out that responsivity to the task preceding tones is altered in comparison to medication naïve participants.

Overall, we showed that sounds acting as warning cues can improve performance in all participants but particularly in ADHD patients. Moreover, and potentially caused by an underlying disorder-specific alertness deficit (Sergeant, 2005), ADHD patients showed increased brain activations during task execution when familiar sounds were presented. We found no evidence for altered processing of novel task preceding sounds, indicating that in this setup ADHD patients showed no higher distractibility than TD peers. Instead, the neural orienting response following novel sounds was unimpaired in the patient group and rIFG activity directly improved task performance.

Thus, our results are encouraging in terms of the identification of optimized learning and working conditions for children and adolescents with ADHD. Their increased susceptibility to beneficial effects of alerting sounds might be used in potential treatments, either by e.g. incorporating external stimulation into homework situations or even by targeting the rIFG and/or left insula with transcranial electrical stimulation. Moreover, the individual reactivity to novelty could potentially be used to identify subtypes of attention profiles in patients with ADHD.

The datasets presented in this article are not readily available because permission to share individual (f)MRI data from the underage study participants was not obtained for ethical reasons. Behavioral data, group fMRI data and code will be made available upon reasonable request from the corresponding author. Requests to access the datasets should be directed to jana.tegelbeckers@ovgu.de.

The study was reviewed and approved by the Local Ethics Committee of the Otto von Guericke University, Faculty of Medicine. Participants and their legal guardian received a detailed oral and written information about the study procedure. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin. Participants were also asked to provide written assent.

JT, AB, BB, H-HF, and KK designed the study. JT and CB-Z performed the experiment. JT and KK analyzed the data and wrote the manuscript. AB, CB-Z, BB, and H-HF edited the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 779, TP A03).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.