Irrelevant stimulus processing in ADHD: catecholamine dynamics and attentional networks

Abstract

A cardinal symptom of attention deficit and hyperactivity disorder (ADHD) is a general distractibility where children and adults shift their attentional focus to stimuli that are irrelevant to the ongoing behavior. This has been attributed to a deficit in dopaminergic signaling in cortico-striatal networks that regulate goal-directed behavior. Furthermore, recent imaging evidence points to an impairment of large scale, antagonistic brain networks that normally contribute to attentional engagement and disengagement, such as the task-positive networks and the default mode network (DMN). Related networks are the ventral attentional network (VAN) involved in attentional shifting, and the salience network (SN) related to task expectancy. Here we discuss the tonic-phasic dynamics of catecholaminergic signaling in the brain, and attempt to provide a link between this and the activities of the large-scale cortical networks that regulate behavior. More specifically, we propose that a disbalance of tonic catecholamine levels during task performance produces an emphasis of phasic signaling and increased excitability of the VAN, yielding distractibility symptoms. Likewise, immaturity of the SN may relate to abnormal tonic signaling and an incapacity to build up a proper executive system during task performance. We discuss different lines of evidence including pharmacology, brain imaging and electrophysiology, that are consistent with our proposal. Finally, restoring the pharmacodynamics of catecholaminergic signaling seems crucial to alleviate ADHD symptoms; however, the possibility is open to explore cognitive rehabilitation strategies to top-down modulate network dynamics compensating the pharmacological deficits.

Keywords: CNV; P300; attention; fMRI; ventral attentional network.

FIGURE 1

The diagram represents the main proposals made in this article. There are four cortical networks regulating behavior, the default mode network (DMN), active at rest; the dorsal attentional Network (DAN), active during task execution and involved in working memory and sustained attention; the ventral attentional network (VAN) involved in attentional shifting; and the salience network (SN) involved in task preparation. The DMN is associated with low tonic catecholaminergic activity, while the DAN is related to moderately high levels of tonic activity. Stimulus-related phasic catecholaminergic signaling activates the VAN, inducing attentional shifts in focused states and transitions between the DAN and the DMN. In addition, phasic signaling activates the SN, which is associated with a buildup of dopaminergic activity related to expectation and task preparation. In ADHD, insufficiently regulated tonic activity results in a disbalanced DMN and distractibility due to a low threshold for phasic signaling, which yields overactivation of the VAN during task performance, and short-term impulsivity. Dysregulation of tonic activity may also result in an attenuated SN due to an incapacity to build up mid-term tonic signaling for task preparation.

FIGURE 2

Lateral (A) and mid-sagittal (B) views of the brain showing the areas involved in the default-mode network (DMN) and the task-positive network. DMN (blue): LPC, lateral parietal cortex; mPFC, middle prefrontal cortex; MTG, mid-temporal gyrus; PC, precuneus; PCC, posterior cingulate cortex. Task-positive areas (yellow/orange): AI, anterior insula; dlPFC, dorsolateral prefrontal cortex; FEF, frontal eye fields; IPL, inferior parietal lobe (anterior aspect); preSMA, pre somatomotor area; PS, precentral sulcus; SMA, somatomotor area. Data from Raichle et al. (2001) and Fox et al. (2005).

FIGURE 3

The dorsal attentional network (DAN) and the ventral attention network (VAN). DAN (yellow/orange): FEF, frontal eye fields; IPS, inferior parietal sulcus; SPL, superior parietal lobe. VAN (blue): IFG, inferior frontal gyrus; IPL, inferior parietal lobe (posterior aspect); MFG, middle frontal gyrus; TPJ, temporo-parietal junction; STG, superior temporal gyrus. Data from Corbetta and Shulman (2002), with permission.

FIGURE 4

Central and peripheral attention in ADHD and control children. (A) We used a face recognition, spatial-shifted double oddball task in which a frequent, standard stimulus (S1) was presented within a central frame, alternating with an infrequent, target stimulus (T1). However, in some trials the frequent or the standard stimuli appeared outside the attentional frame (S2, T2, respectively). (B) P300 ERPs elicited by the distinct stimuli in a central ROI. Thick continuous line: S1, standard stimulus on attentional focus. Thin continuous line: T1, target stimulus on attentional focus. Dotted line: T2, target stimulus in the periphery. Segmented line: S2, standard stimulus in the periphery. The results indicate that in the focused condition, only the target stimulus (T1) elicits a P300, both in ADHD and controls. However, for peripheral stimuli, a P300 deflection is observed only in ADHD. For further details, see López et al. (2006) with permission.

FIGURE 5

The attentional blink in ADHD and controls. (A) The task consists of a rapid serial visual presentation paradigm in which there is a salient, colored letter (T) followed by several letters, among them an X which may or may not be present, at different distances from the “T.” Detection of the second letter (“X”) is minimal when it is presented, as shown in the figure, some 80 ms after the “T.” (B) ERPs produced by the second letter (“X”) during the attentional blink. ERPs show a morphology that is characteristic of steady-state evoked potentials. After subtracting the effect of the P300 elicited by the first letter (“T”), a P300 was present in control subjects only when they detected the second letter (“X”; green dotted line). If this was present but unnoticed, there was no P300 elicited (red solid line). In ADHD, P300 were of significantly smaller magnitudes [F_(1,22) = 17.64, p < 0.01], as expected, but were present both when the second letter (“X”) was detected, and when it passed unnoticed (note the magnification of voltage scale in the ADHD group, in order to visualize better the differences between conditions). When a letter different from “X” was present, there was no P300 in either group (blue line). The P300 observed displays a longer latency than usual due to the experimental design (a previous relevant stimulus and a steady sate visual ERP paradigm, which produce an increased delay in the stimulus-driven response). Between-group comparisons of P300 latency and topography were not statistically significant. For further details, see López et al. (2008) with permission.

FIGURE 6

Preparatory activity during a visual search task in ADHD and controls. (A) The subject had to fixate in front a cross flanked by four squares. Subsequently, one (low attentional load) or two (high attentional load) of these squares lighted in a different color in order to deviate attention. After this, an array of zeroes was presented inside the squares, among which there was also an “X” that had to be detected. (B) ERPs elicited during the task. The array of zeroes in the frame was presented at about 800 ms after the presentation of the cue frames. Between 300 and 800 ms, a contingent negative variation (CNV) potential develops, which is indicative of expectancy for the task. Note that in ADHD the CNV is noticeably lower than in controls. No differences in the CNV were seen between the high and low attentional load conditions. For further details, see Ortega et al. (2013) with permission. **p < 0.005; ***p < 0.001.