حواس پرتی عاطفی در پسران مبتلا به ADHD: ارتباطات عصبی و رفتاری

Abstract Although, in everyday life, patients with attention deficit hyperactivity disorder (ADHD) are frequently distracted by goal-irrelevant affective stimuli, little is known about the neural and behavioral substrates underlying this emotional distractibility. Because some of the most important brain responses associated with the sudden onset of an emotional distracter are characterized by their early latency onset and short duration, we addressed this issue by using a temporally agile neural signal capable of detecting and distinguishing them. Specifically, scalp event-related potentials (ERPs) were recorded while 20 boys with ADHD combined type and 20 healthy comparison subjects performed a digit categorization task during the presentation of three types of irrelevant, distracting stimuli: arousing negative (A−), neutral (N) and arousing positive (A+). Behavioral data showed that emotional distracters (both A− and A+) were associated with longer reaction times than neutral ones in the ADHD group, whereas no differences were found in the control group. ERP data revealed that, compared with control subjects, boys with ADHD showed larger anterior N2 amplitudes for emotional than for neutral distracters. Furthermore, regression analyses between ERP data and subjects’ emotional ratings of distracting stimuli showed that only in the ADHD group, emotional arousal (ranging from calming to arousing) was associated with anterior N2: its amplitude increased as the arousal content of the visual distracter increased. These results suggest that boys with ADHD are more vulnerable to the distracting effects of irrelevant emotional stimuli than control subjects. The present study provides first data on the neural substrates underlying emotional distractibility in ADHD.

1. Introduction The ability to remain goal oriented in the face of irrelevant distracting stimuli is crucial for successful adaptive functioning. This ability is thought to depend on two closely interrelated and mutually dependent attentional mechanisms (Corbetta et al., 2008 and Corbetta and Shulman, 2002). On the one hand, voluntary top-down processes are triggered and developed by knowledge, expectation and current goals (e.g., read a book for an exam). On the other hand, involuntary bottom-up processes are driven by stimulus features such as novelty or significance (e.g. a wasp that appears suddenly while reading the book). Interestingly, emotional stimuli, salient and signal events by definition, have been shown to be prominent distracters that can efficiently capture attention in a bottom-up fashion, thereby disrupting the focus on goal-relevant information (Carretié et al., 2004, Carretié et al., 2005, Vuilleumier and Schwartz, 2001 and Öhman et al., 2001). An increased susceptibility to distraction is currently one of the behavioral diagnostic criteria of attention-deficit/hyperactivity disorder (ADHD; American Psychiatric Association, 2000). Indeed, the presence of heightened levels of distraction in ADHD is believed to be associated with broad impairment across multiple domains, including cognitive functioning (e.g., disrupting the ability to maintain information in working memory: Higginbotham and Bartling, 1993 and Marx et al., 2011), interpersonal relationships (e.g., making difficult to follow the sequence of rules in social activities: Maedgen & Carlson, 2000), academic or work performance (e.g. making careless mistakes in school or job activities: Shifrin, Proctor, & Prevatt, 2010), and health (e.g., increasing distraction-related accidents and associated injuries: Barkley & Cox, 2007). However, experimental evidence of enhanced distractibility in ADHD is equivocal. Whereas some behavioral and electrophysiological data suggest that individuals with ADHD are more distractible than healthy comparison subjects (Gumenyuk et al., 2005, Mason et al., 2005, Radosh and Gittelman, 1981 and Rosenthal and Allen, 1980), others have reported that patients with this disorder are not affected by irrelevant distracting stimuli to a greater extent than controls (Booth et al., 2005, Huang-Pollock et al., 2005, Jonkman et al., 2000 and Meere and Sergeant, 1988). A recent study has even shown that, in certain circumstances, the presence of auditory distracters could improve the performance of children with ADHD (van Mourik, Oosterlaan, Heslenfeld, Konig, & Sergeant, 2007). In any case, it should be mentioned that research on this topic is scarce, particularly in comparison with the large body of data on the neural mechanisms underlying the reduced top-down inhibitory control in ADHD (Albrecht et al., 2008, Dimoska et al., 2003, Liotti et al., 2005, Pliszka et al., 2000 and Rubia et al., 1999), which has been traditionally proposed as the core deficit of this disorder (Barkley, 1997). However, growing evidence indicates that this deficit in inhibitory control is not present among all patients with ADHD and, in some cases, is preceded and caused by other processing deficits (Banaschewski et al., 2004, Brandeis et al., 1998, McLoughlin et al., 2010 and Willcutt et al., 2005). This evidence has led to question whether inhibition is the central deficit in ADHD and to look for the involvement of other psychopathological processes, including bottom-up and affective mechanisms (Castellanos et al., 2006, Nigg and Casey, 2005, Sergeant, 2005 and Sonuga-Barke, 2002). It should be also noted that previous studies on attentional deficits in ADHD have relied heavily on emotionally neutral visual distracters, such as letters, numbers and geometric shapes (Booth et al., 2005, Huang-Pollock et al., 2005, Jonkman et al., 2000 and Mason et al., 2005). In real social situations, however, maintaining goal-directed attention in the face of salient affective distracters is often needed. Convergent evidence from hemodynamic and electrophysiological studies suggests enhanced neural responses to emotional stimuli relative to neutral ones, even when these stimuli are not consciously perceived (Carretié et al., 2005, Vuilleumier and Schwartz, 2001 and Whalen et al., 1998). For example, a number of investigations have reported amplified responses to emotional visual events, involving structures such as the amygdala and the extrastriate visual cortex as well as early and late electrophysiological responses such as N2 and P3 (see reviews by Olofsson et al., 2008 and Vuilleumier, 2005). Therefore, employing emotional stimuli may help to evoke clearer distraction effects in conditions simulating real social environments. Furthermore, the idea of incorporating emotional stimuli in the characterization of ADHD fits well with current models that emphasize that multiple psychopathological processes and neural pathways are implicated in this disorder, including cognitive (e.g., attention, inhibition and working memory) and affective (emotion and motivation) processes as well as top-down (voluntary) and bottom-up (involuntary) mechanisms (Castellanos et al., 2006, Nigg and Casey, 2005, Sergeant, 2005 and Sonuga-Barke, 2002; see also Sonuga-Barke, De Houwer, De Ruiter, Ajzenstzen, & Holland, 2004). From this perspective, the poor ability of ADHD patients to remain focused on a task in the presence of irrelevant emotional distracters could arise not only from a hypofunction of the brain processes associated with cognitive control of distraction, but also from a hyperfunction of brain processes related to the bottom-up response to affectively laden stimuli. In support of this, a recent fMRI study has shown that adolescents with ADHD displayed amygdalar hyperactivity during subliminal presentation of fearful faces (Posner et al., 2011b). However, to the best of our knowledge, no study has yet addressed the effect of emotional irrelevant stimuli on ongoing cognitive processes in children with ADHD. Due to their high temporal resolution that allows neural processes to be tracked in milliseconds, event-related potentials (ERPs) are particularly useful for elucidating the neural basis underlying emotional distraction in ADHD. The main reason for this is that some of the most important brain responses associated with the sudden onset of an emotional distracter are characterized by their rapidity (early latency onset) and brevity (short duration), and thereby can only be detected by using a temporally agile physiological signal such as electroencephalography (EEG). One ERP component that seems particularly well suited for studying emotional distractibility in the visual modality is the anterior N2, a brain electrical response occurring between 200 and 400 ms after stimulus onset that presents its maximum amplitude over frontal scalp regions. Numerous studies have shown this component to be enhanced for unfamiliar, novel visual stimuli as well as for highly emotional events (Carretié et al., 2004, Chong et al., 2008 and Daffner et al., 2000; Kenemans, Verbaten, Melis, & Slangen, 1992; Liddell et al., 2004 and Rozenkrants and Polich, 2008). Remarkably, it has recently been reported that, unlike subsequent positive components, the amplitude of the anterior N2 to this type of stimuli is neither modulated by the degree of task-relevance of the eliciting stimulus nor by the direction of subjects’ controlled attention (Chong et al., 2008 and Tarbi et al., 2011; see also Carretié et al., 2004 and Liddell et al., 2004). Therefore, this component responds to novel and emotional events even when they are not relevant for the task and occur outside the focus of attention. In light of this evidence, this anterior N2, which is thought to be functionally distinct from the frontocentrally distributed control-related N2 mainly elicited by executive control paradigms (see Folstein & Van Petten, 2008 for a review on this issue), seems to reflect automatic detection of highly significant stimuli (Chong et al., 2008, Daffner et al., 2000, Liddell et al., 2004 and Tarbi et al., 2011). To our knowledge, no study has examined it in ADHD. Following the anterior N2, a large positive deflection over centro-parietal regions is often observed. This posteriorly distributed positivity has been variously called P3, P3b, LPP or LPC, and has generally been associated with more controlled stages of processing (Chong et al., 2008, Kenemans et al., 1992 and Liddell et al., 2004). For instance, P3b is thought to reflect the processing of task-relevant events, including stimulus categorization/evaluation and memory updating (Donchin, 1981, Kok, 2001, Polich, 2007 and Verlerger, 1998). ERP studies of patients with ADHD have frequently shown a reduction in the amplitude of P3b to task-relevant stimuli (Barry et al., 2003, Brandeis et al., 2002 and Jonkman et al., 2000). Within the context of emotion research, this component (often termed LPP) has been shown to be sensitive to manipulations requiring voluntary control processes. Indeed, it has been been proposed as a neural marker of top-down emotion regulation in both adults and children (Dennis and Hajcak, 2009 and Moser et al., 2006). Interestingly, a reduced amplitude of LPP has recently been found in patients with ADHD when they asked to inhibit their responses to negative emotions (Köchel, Leutgeb, & Schienle, 2012). Hereafter, we will use the term late positive complex (LPC) to describe this family of posteriorly distributed positivities associated to a greater extent than previous components with controlled and conscious processes. The present study aimed at elucidating the neural and behavioral mechanisms underlying emotional distraction in children with ADHD. To this end, ERP and behavioral data were recorded from boys with ADHD combined type and healthy comparison controls while they performed a digit categorization task while three types of irrelevant, distracting stimuli were presented: arousing negative (A−), neutral (N) and arousing positive (A+). Specifically, behavioral measures consisted of reaction times (RTs) and error rates in the cognitive task. Distraction caused by the irrelevant emotional stimuli would be mirrored in an impoverishment of current task performance (i.e., longer RTs and/or higher error rates for emotional versus neutral distracters). Neural measures consisted of scalp ERP analyses of N2 and LPC. In this case, emotional distraction would be reflected in alterations of one or both of these components. An enhancement in the anterior N2 would suggest that distraction is primarily related to an exaggerated bottom-up response to the saliency of distracters, whereas a reduction in the LPC would indicate that distraction is associated with deficits in a later, more controlled stage of processing. On the basis of evidence showing heightened distractibility in children and adolescents with ADHD when they are in real social situations (Barkley and Cox, 2007, Lawrence et al., 2002 and Lorch et al., 2000), we hypothesized that boys with ADHD may show enhanced susceptibility to emotional distraction as compared to matched control subjects, both at the behavioral and electrophysiological levels.

. Results 3.1. Emotional assessments Table 2 shows the means and standard deviations of valence and arousal for each group and each type of distracter. As previously described, ANOVAs were computed for both emotional dimensions, using Group as between-subjects factor and Distracter type as within-subjects factor. The main effect of Distracter type was significant both for valence (F(2, 76) = 634.5, p = 0.000, ε = 0.82, View the MathML sourceηp2=0.94) and arousal dimensions (F(2, 76) = 257.6, p = 0.000, ε = 0.94, View the MathML sourceηp2=0.87). Bonferroni post hoc contrasts indicated that A− and A+ showed different valences (p = 0.000) but not different arousal levels (p = 0.091). Moreover, both A+ and A− differed from N in arousal (A+ vs. N, p = 0.000; A− vs. N, p = 0.000) and valence (A+ vs. N, p = 0.000; A− vs. N, p = 0.000). The main effect of Distracter type remained significant after statistically controlling for age, both for valence (F(2, 74) = 13.4, p = 0.000, ε = 0.82) and arousal (F(2, 74) = 7.6, p = 0.001, ε = 0.94). Present results therefore confirm that the affective valence of each distracter type was as assumed a priori, and that negative and positive distracters were balanced with respect to their arousal levels. Importantly, the interaction between Group and Distracter type were not significant neither for valence (F(2, 76) = 0.7, p = 0.466, ε = 0.82) nor for arousal (F(2, 76) = 1.3, p = 0.283, ε = 0.94). These findings indicate that there were no differences between ADHD and healthy control groups in the subjective feeling of valence and arousal caused by each distracter type. Table 2. Group means and standard deviations on subjective emotional ratings as well as on behavioral and electrophysiological (anterior N2) measures for each distracter type (A−, arousing negative; N, neutral; A+, arousing positive). ADHD group Control group A− N A+ A− N A+ Valencea 1.95 (0.41) 3.09 (0.14) 4.43 (0.28) 1.74 (0.31) 3.05 (0.18) 4.3 (0.39) Arousalb 4.09 (0.42) 2.95 (0.14) 4.36 (0.25) 4.19 (0.35) 2.96 (0.17) 4.24 (0.4) RTsc 704.5 (175.45) 668.36 (155.59) 696.29 (164.31) 807.26 (170.76) 792.71 (172.02) 794.62 (169.21) Error ratesd 0.13 (0.1) 0.13 (0.1) 0.15 (0.12) 0.1 (0.08) 0.09 (0.09) 0.1 (0.06) Anterior N2 factor scoree −0.43 (0.78) 0.36 (0.81) −0.65 (0.95) 0.22 (1.02) 0.43 (0.77) 0.07 (1.19) Anterior N2 amplitudef −4.19 (2.39) −2.21 (2.26) −4.9 (2.77) −2.73 (2.68) −2.32 (2.09) −3.09 (3.28) a Valence: subjects’ valence ratings of distracting stimuli (from 1, negative, to 5, positive). b Arousal: subjects’ arousal ratings of distracting stimuli (from 1, negative, to 5, positive). c RTs: reaction times (in ms). d Error rates: omissions and incorrect button presses, divided by the number of trials. e Anterior N2 factor score: as explained in detail in the text, a single parameter that reflects the mean amplitude of the whole spatial factor. Negative values respresent higher amplitudes. f Anterior N2 amplitude: average voltage (in μV) occurring in the 200–400 ms time interval across six frontocentral electrodes (see Supplementary Fig. 1). Table options 3.2. Behavioral data Performance in the digit categorization task is shown in Table 2. Repeated-measures ANOVAs on RTs and error rates were performed using Group and Distracter type as factors. The main effect of Group was not significant for error rates (F(1, 38) = 0.8, p = 0.381). In the case of RTs, the main effect of Group was marginally significant (F(1, 38) = 4.2, p = 0.047, View the MathML sourceηp2=0.1), showing that RTs were shorter in the ADHD than in the control group (means and standard errors: 689.72 ± 37.37 and 798.2 ± 37.37, respectively). The interaction of Group and Distracter type was not significant in the case of error rates (F(2, 76) = 0.2, p = 0.789, ε = 0.9), but produced significant effects in RTs (F(2, 76) = 4.4, p = 0.021, ε = 0.85, View the MathML sourceηp2=0.1). Post hoc tests of simple effects with Bonferroni correction for multiple comparisons showed that emotional distracters (both A+ and A−) were associated with longer RTs than neutral distracters in the ADHD group (A− vs. N, p = 0.000; A+ vs. N, p = 0.000; A− vs. A+, p = 0.717), whereas no differences among distracters were observed in the Control group (A− vs. N, p = 0.195, A+ vs. N, p = 1, A+ vs. A−, p = 0.220). This interaction remained significant after controlling for age (F(2, 74) = 4.8, p = 0.015, ε = 0.86, View the MathML sourceηp2=0.11). Age was significant as a covariate (F(1, 37) = 20.1, p = 0.000, View the MathML sourceηp2=0.35). Specifically, RTs decreased linearly with age (r = −0.64, p = 0.000). Group effects, however, were no longer significant when age was added as a covariate (means and standard errors for ADHD and Control groups: 715.46 ± 31.02 and 772.46 ± .31.02, respectively; F(1, 37) = 1.6, p = 0.209). Therefore, although age differences between groups were not significant ( Table 1), the fact that the children with ADHD were somewhat older than control children seems to be the reason for the group differences in RTs. 3.3. ERP data Fig. 2 shows a selection of grand averages once the baseline value (pre-stimulus recording) had been subtracted from each ERP. Two main deflections were noticeable. The first was a negative component with an onset around 280 ms after distracter presentation (N2). This component shows its maximum amplitude over frontal and frontocentral electrode locations. The second deflection was a late positivity peaking around 585 ms after distracter onset (LPC). The largest amplitude of this positivity was observed at centroparietal locations. Fig. 3 represents the distribution of voltages of these components in the form of scalp maps for the ADHD and Control groups. Grand average ERP response to each distracter type (A−, Arousing negative; N, ... Fig. 2. Grand average ERP response to each distracter type (A−, Arousing negative; N, neutral; A+, Arousing positive) for the ADHD and control groups. Grand average waveforms are shown at frontal and centroparietal electrodes, where N2 and LPC were more prominent, respectively. Figure options Topographic maps of the voltage distribution of N2 and LPC (collapsing across ... Fig. 3. Topographic maps of the voltage distribution of N2 and LPC (collapsing across distracter types) for the ADHD and control groups. Red represents positive values and blue represents negative values (μV). Because N2 is a negative wave, blue indicates larger N2 amplitudes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.). Figure options 3.3.1. Temporospatial PCA As a consequence of the application of the tPCA using the PA as criterion of the number of factor to retain (Fig. 4A), five components were extracted from the ERPs (Fig. 5). Factor peak latency and topography characteristics associate Factor 1 (peaking at 280.95 ms) with the wave labeled N2 in grand averages and Factor 2 (peaking at 585.71) with that labelled LPC. These labels will be employed hereafter to make the results easier to understand. The sPCA subsequently applied to N2 and LPC temporal factor scores extracted three spatial factors for N2 and two for LPC (Fig. 6). The decision on the number of spatial factors to select for these two ERP components was also based on the PA (Fig. 4B) Parallel Analysis to determine the number of factors to retain in tPCA (A) and ... Fig. 4. Parallel Analysis to determine the number of factors to retain in tPCA (A) and sPCAs (B): Plots of eigenvalues from the real and random datasets. Only those factors from the real datasets that explained more variance (i.e., had greater eigenvalues) than corresponding factors in the simulated random datasets were retained. Figure options Temporal principal component analysis (tPCA): factor loadings after Promax ... Fig. 5. Temporal principal component analysis (tPCA): factor loadings after Promax rotation. Temporal factors 1 (66.14% explained variance) and 2 (16.52% explained variance) are drawn in black. As mentioned in the text, only temporal factor 1 (N2) was sensitive to experimental manipulations. Figure options Spatial factors extracted for N2 (A) and LPC (B) through spatial principal ... Fig. 6. Spatial factors extracted for N2 (A) and LPC (B) through spatial principal component analysis (sPCA). An asterisk signals the spatial factor sensitive to the experimental manipulations (anterior N2). This spatial factor accounted for 54.79% of the variance of N2. Figure options Repeated-measures ANOVAs on N2 and LPC spatial factor scores (directly related to amplitudes, as previously indicated) were then carried out for Group and Distracter type factors. The main effect of Group was not significant in any of the N2 spatial factors. In other words, amplitudes did not differ between groups in the anterior N2 (F(1, 38) = 3.3, p = 0.079), centroparietal N2 (F(1, 38) = 0.06, p = 0.803) or occipitally distributed N2 (F(1, 38) = 0.05, p = 0.821). However, a significant effect of the interaction of Group and Distracter type was obtained in the anterior spatial factor of N2 (F(2, 76) = 5.7, p = 0.006, ε = 0.91, View the MathML sourceηp2=0.13). Specifically, post hoc tests of simple effects with Bonferroni correction for multiple comparisons showed that anterior N2 amplitudes were larger (more negative) for emotional distracters (both A+ and A−) than for neutral distracters in the ADHD group (A− vs. N, p = 0.000; A+ vs. N, p = 0.000; A+ vs. A−, p = 0.324), whereas no differences were observed in the Control group (A− vs. N, p = 0.407, A+ vs. N, p = 0.124, A+ vs. A−, p = 0.853). This interaction remained significant after controlling for age (F(2, 74) = 6.9, p = 0.002, ε = 0.92, View the MathML sourceηp2=0.16). Age was significant as a covariate (F(1, 37) = 5.3, p = 0.027, View the MathML sourceηp2=0.12). Concretely, the amplitude of the anterior N2 decreased linearly with age (r = 0.21, p = 0.022), which is consistent with previous studies showing age-related changes in the anterior N2 ( Riis et al., 2009 and Van Strien et al., 2011). Neither the main effect of Group nor the interaction of Group and Distracter type was significant in the centroparietally distributed LPC (Group: F(1, 38) = 0.7, p = 0.395; Group x Distracter type: F(2, 76) = 0.3, p = 0.754, ε = 0.99) or the anteriorly distributed LPC (Group: F(1, 38) = 0.2, p = 0.695; Group x Distracter type: F(2, 76) = 0.7, p = 0.516, ε = 0.98). 3.3.2. Conventional ERP analysis In order to compare results from temporospatial PCA with traditional ERP waveform analysis, we also assessed the N2 and LPC as mean voltage amplitudes within the 200–400 ms and 500–700 ms intervals, respectively. Amplitudes were measured with respect to the average of the 200 ms pre-stimulus baseline. Scalp regions of interest were defined, and the average amplitude recorded by those electrodes forming each of these regions was computed. Time windows and scalp regions of interest were determined based on previous research as well as on visual inspection of grand averages and topographic map of the voltage distribution of each component (Fig. 2 and Fig. 3). For N2, a frontocentral region comprising six electrodes (AFz, F3, Fz, F4, FC1, and FC2) was selected. For LPC, a centroparietal region comprising five electrodes (CP1, CP2, P3, Pz, and P4) was chosen. Supplementary Fig. 1 shows the selected electrode region for each component. Repeated-measures ANOVAs on these two regions were then carried out for Group and Distracter type factors. Results were similar to those obtained using temporospatial PCA. For anterior N2, the main effect of Group was not significant (F(1, 38) = 2.1, p = 0.152). However, it was sensitive to the interaction of Group and Distracter type (F(2, 76) = 4.4, p = 0.016, ε = 0.97, View the MathML sourceηp2=0.1). Bonferroni post hoc tests of simple effects showed that anterior N2 amplitudes were larger for emotional (both A− and A+) than for neutral distracters in the ADHD group (A− vs. N, p = 0.000; A+ vs. N, p = 0.000; A+ vs. A−, p = 0.441), but not in the Control group (A− vs. N, p = 1; A+ vs. N, p = 0.447; A+ vs. A−, p = 1). This interaction remained significant after controlling for age (F(2, 74) = 5.9, p = 0.005, ε = 0.98, View the MathML sourceηp2=0.14). Neither the main effect of Group (F(1, 38) = 0.01, p = 0.902) nor the interaction of Group and Distracter type (F(2, 76) = 0.5, p = 0.621, ε = 0.87) was significant in the centroparietally distributed LPC. 3.4. Relationship between emotional assessments and ERP data Although it is reasonable to deduce from previous analyses that arousal more than valence explains results concerning the anterior N2 component in the ADHD group, since differences between emotional (both A+ and A−) and neutral distracters are clear, this trend required statistical confirmation. To this end, the associations between anterior N2 factor scores (equivalent to traditional amplitudes, as mentioned before) and subjects’ emotional ratings of distracting stimuli were analyzed via multiple regressions (enter method) in both groups. Anterior N2 amplitude was the dependent variable, and independent variables were valence and arousal ratings. In the ADHD group, arousal associated significantly with anterior N2 (F(2, 57) = 6.02, p = 0.004; β = −0.4, p = 0.002), while valence did not (β = −0.09, p = 0.475). Concretely, the linear association pattern between anterior N2 amplitudes and arousal presented a negative slope: the larger -more negative- the former, the greater the latter. By contrast, neither valence nor arousal was associated with anterior N2 amplitude in the control group (F(2, 57) = 0.3, p = 0.735; β = −0.06, p = 0.657 and β = −0.08, p = 0.531, respectively). Table 3 shows, for each group, the correlations between the independent variables, and between each independent variable and the dependent variable. The inclusion of age as an additional predictor variable did not alter the results of the regression analyses, neither for the ADHD nor for the control group. Table 3. Correlations between variables included in the regression model for the ADHD and control groups (dependent variable: anterior N2 amplitudes; independent variables: subjective valence and arousal ratings). ADHD group Control group Anterior N2 spatial score Valence Arousal Anterior N2 spatial score Valence Arousal Anterior N2 spatial scorea – – Valenceb rp = −0.1 – rp = −0.06 – p = 0.47 p = 0.66 Arousalc rp = −0.4 r = 0.14 – rp = −0.08 r = 0.04 – p = 0.002 p = 0.28 p = 0.53 p = 0.78 r = Pearson correlation; rp = Partial Pearson correlation, controlling for valence/arousal. a Anterior N2 factor score: a single parameter that reflects the mean amplitude of the whole spatial factor. b Valence: subjects’ valence ratings of distracting stimuli (from 1, negative, to 5, positive). c Arousal: subjects’ arousal ratings of distracting stimuli (from 1, calming, to 5, arousing).