شواهد الکتروفیزیولوژیک و تفاوت های جنسیتی رفتاری در مدولاسیون حواس پرتی توسط زمینه عاطفی

Abstract Gender differences in brain activity while processing emotional stimuli have been demonstrated by neuroimaging and electrophysiological studies. However, the possible differential effects of emotion on attentional mechanisms between women and men are less understood. The present study aims to elucidate any gender differences in the modulation of unexpected auditory stimulus processing using an emotional context elicited by aversive images. Fourteen men and fourteen women performed a well-established auditory–visual distraction paradigm in which distraction was elicited by novel stimuli within a neutral or negative emotional context induced by images from the IAPS. Response time increased after unexpected novel sounds as a behavioral effect of distraction, and this increase was larger for women, but not for men, within the negative emotional context. Novelty-P3 was also modulated by the emotional context for women but not for men. These results reveal stronger novelty processing in women than in men during a threatening situation.

1. Introduction Emotional disorders bring great distress to human life and are known to have an asymmetrical prevalence among men and women. For example, the prevalence of major depressive disorder in women is 5–9%, while in men it is only 2–3%. Similarly, anxiety disorders are much more frequently diagnosed in women than in men (DSM-IV-R, 1997). Consequently, gender differences in emotional processing have been the object of recent research due to their importance for understanding the unequal prevalence of anxiety and emotion-related disorders in women and men. Previous studies have demonstrated that men and women process emotional stimuli differently. For instance, Kemp et al. (2004) recorded steady-state, visually evoked potentials elicited to pictures from the International Affective Picture System (IAPS; Lang et al., 2005). They found widespread frontal latency reduction associated with the processing of unpleasant pictures in women as compared to men, suggesting that women were more responsive to emotionally loaded stimuli. In an event-related potential (ERP) study, Orozco and Ehlers (1998) found significantly enhanced amplitudes in frontal regions in response to sad faces as compared to neutral ones in both men and women; however, the amplitude was larger for women in comparison to men on the P450, an ERP component related to the “intensity” of response to emotional stimuli, thus suggesting that women may be more “sensitive” to emotional stimuli. Moreover, neuroimaging studies exploring the neural correlates of emotional processing have shown stronger right amygdala activation in women as compared to men (Wrase et al., 2003 and Hofer et al., 2006), and stronger activation of its related cerebral network (Canli et al., 2002). The amygdala, a subcortical structure located in the anterior medial temporal lobe, has a crucial role in the processing of emotions (LeDoux, 2000), and it has been shown to differ in men and women not only functionally but also in terms of structure and in several aspects of its ontogenesis (Goldstein et al., 2001). On the other hand, research has also shown that negative emotions interact with the processing of concomitant stimuli, either facilitating or competing with them; this occurs not only within (Mogg and Bradley, 1999 and Anderson, 2005) but also across sensory modalities. For instance, stimuli with an affective load have been shown to elicit stronger and faster attention capture than do non-emotional stimuli (Hansen and Hansen, 1988 and Öhman et al., 2001). Emotional stimuli capture attention at very early stages of information processing in the human brain, circa 100–200 ms from stimulus onset (Smith et al., 2003, Carretié et al., 2004 and Krolak-Salmon et al., 2004) or earlier, around 50–100 ms from stimulus onset (Sugase et al., 1999), even when the emotional stimuli occur outside the attentional focus, that is, automatically (Vuilleumier et al., 2001, Pessoa et al., 2002, Schupp et al., 2003, Carretié et al., 2005 and Williams et al., 2004). A visually induced negative emotion interacts with the processing of concomitant auditory inputs, as indexed by startle reflex potentiation via picture arousal (Amrhein et al., 2004 and Stanley and Knight, 2004) and P3 attenuation through the valence of the image (Schupp et al., 1997 and Cuthbert et al., 1998). Moreover, a recent fMRI study has reported that areas involved in auditory novelty processing (bilateral superior temporal gyri) are significantly more activated in a negative emotional context as compared with a neutral one (Domínguez-Borràs et al., 2008b), thus demonstrating how an emotional context can modulate the orientation of attention towards salient stimuli, analogous to other top-down mechanisms of attentional control (SanMiguel et al., 2008). In a similar ERP study the electrophysiological results revealed an enhancement of novelty-P3 when novel sounds were processed during the performance of a task involving emotionally negative pictures (Domínguez-Borràs et al., 2008a). The study of gender differences for such emotional modulation of involuntary attention mechanisms may help to elucidate the different impact that an emotional environment exerts on women and men. In order to examine the possible differential modulatory effects of emotion on involuntary attention we used a modified version of a well-characterized “distraction paradigm” (see reviews in Escera et al., 2000, Escera and Corral, 2003 and Escera and Corral, 2007), derived from an adaptation of the so-called “oddball” paradigm, in which a high-probability standard stimulus is replaced randomly by a rare or “odd” stimulus. In the auditory–visual distraction paradigm a sequence of task-irrelevant frequent standard sounds (e.g. 80%) and infrequent deviant or novel sounds (e.g. 20%) is delivered. Task-relevant targets were digits presented 300 ms after the to-be-ignored auditory stimuli, and participants were instructed to classify them (e.g. as odd or even). This timing is optimal for eliciting distraction and accurately measuring behavioral and brain responses related to involuntary auditory attention (Escera et al., 2000, Escera and Corral, 2003 and Escera and Corral, 2007); it was chosen on the basis of a previous study that showed distracting effects of deviant sounds when they occurred 200 ms before the target but not when they occurred 560 ms in advance (Schröger, 1996). Behavioral distraction has been shown through delayed response times obtained in a range of experiments using both auditory–auditory (Schröger and Wolff, 1998, Bert and Schröger, 2001 and Roeber et al., 2003) and auditory–visual paradigms (Alho et al., 1997, Escera et al., 1998, Escera et al., 2000, Escera et al., 2001, Escera et al., 2002, Escera et al., 2003, Yago et al., 2001a, Yago et al., 2001b and Yago et al., 2003). Concomitant recordings of ERPs show the “distraction potential”, a well-defined electrophysiological response which begins with a negative deflection combining an enhancement of the auditory N1 component and the mismatch negativity (MMN). This early negativity indexes the mechanisms for stimulus change detection, leading to attention capture ( Escera et al., 1998 and Alho et al., 1998). The auditory N1 component is elicited by the onset of any abruptly commencing sound ( Näätänen and Picton, 1987), and the subsequent MMN is an ERP component elicited by any discriminable change in the otherwise repetitive auditory stimulation, reflecting automatic change detection ( Näätänen, 1989 and Näätänen et al., 2007). N1-enhancement/MMN is followed by a prominent positivity, the so-called novelty-P3 or P3a, which is associated with the evaluation of these novel events for subsequent behavioral action ( Friedman et al., 2001) and reflects an effective orienting of attention toward the detected change ( Friedman et al., 2001 and Escera et al., 1998). P3a is an ERP component of the “P300 family”, which discloses the early and frontocentral P3a elicited by unexpected stimuli from the later centro-parietal P3b elicited to task-relevant target stimuli (see Polich, 2007). The P3a or novelty-P3 component has been described to have two subcomponents, the early and late ones, which are clearly distinguished on the basis of their respective latency, scalp distribution and psychological concomitants ( Escera et al., 1998, Escera et al., 2000, Polo et al., 2003, Yago et al., 2003, SanMiguel et al., 2008 and Cortiñas et al., 2008). After the novelty-P3, another negative response may be observed: this is the ‘reorienting negativity’ (RON), reflecting processes in the context of reorienting attention towards task-relevant aspects of stimulation following distraction ( Schröger and Wolff, 1998). These typical waveforms index three main stages of exogenous attentional control: acoustic change detection, effective orienting of attention towards the detected change, and reorienting of attention towards the current task ( Escera et al., 2000 and Escera and Corral, 2003). It has been suggested that the processing of novel sounds may be modulated if the amount of attentional resources available is modified by the task conditions (Berti and Schröger, 2003, SanMiguel et al., 2008 and Lavie, 2005), for instance, by increasing working memory load. The auditory–visual distraction paradigm has been demonstrated to be a useful tool for examining these effects of task conditions on the mechanisms of auditory novelty processing and distraction (SanMiguel et al., 2008, Domínguez-Borràs et al., 2008a and Domínguez-Borràs et al., 2008b), and it may therefore be suitable for investigating the impact of emotions on attentional processing and the potential gender differences of this impact. Unexpected novel stimuli in an emotionally negative situation, such as a threatening or fear environment, acquire a vital importance as they may be potentially harmful; therefore, stronger processing becomes crucial and has an obvious adaptive value. The present study aimed to elucidate by means of ERPs any gender differences in the modulatory effect of an aversive emotional context on the brain response to task-irrelevant novel sounds. As previous studies have demonstrated that women have a more intensive response to affective stimuli, we hypothesised that a negative emotional context would have a larger modulatory effect on auditory novelty processing in female compared to male participants.

Results 3.1. Behavioral results Participants had an overall hit rate of about 91%, which decreased significantly in the emotionally negative condition (Context: F(1,27) = 8.95, p = 0.006), as well as on trials including a novel sound (Sound: F(1,27) = 5.10, p = 0.032; see Fig. 1). A two-factor repeated-measures ANOVA revealed that response time was significantly delayed after novel sounds (Sound: F(1,27) = 41.36, p < 0.001), indicating that subjects were distracted by the unexpected occurrence of task-irrelevant novel sounds, in agreement with previous studies ( Alho et al., 1997, Escera et al., 1998, Escera et al., 2001 and Escera et al., 2003). Response time was also significantly longer in the negative condition (Context: F(1,27) = 15.12, p = 0.001). Although there was no significant interaction between these two factors, a trend toward longer response delay after novel sounds was apparent in the negative context as compared to the neutral context (Sound × Context: F(1,27) = 2.99, p = 0.095). Further analysis for male and female subjects separately indicated that this marginal effect could be due to women's performance, where the distraction effect on response time was significantly larger in the emotionally-negative context (Sound × Context: F(1,14) = 7.85, p = 0.014), while in men no significant interaction between context and sound was found. No statistical differences were observed between counterbalanced runs or response buttons for either RT or hit rate. Visual task performance for women and men in standard and novel trials and for ... Fig. 1. Visual task performance for women and men in standard and novel trials and for the two emotional conditions. Response times (A) increased and hit rate (B) decreased for both men and women after novel auditory stimuli and for negative emotional context. Only in women the distraction effect elicited by novel sounds increased in the negative emotional context condition. Figure options 3.2. Electrophysiological results 3.2.1. Negative versus neutral stimulus comparison Grand-average ERPs to standard trials for both conditions (see Section 2) revealed a sequence of auditory P1, N1 and P2 and visual target-related P300 deflections (Fig. 2). The statistical scalp-distribution analyses of target-P300 across neutral and negative conditions (ANOVA for normalized amplitudes with three factors: Context, Laterality and Frontality) yielded a P300 distribution over posterior electrodes in the four consecutive latency windows (Frontality: F(4,104) = 211.59, p < 0.001; F(4,104) = 261.15, p < 0.001; F(4,104) = 72.75, p < 0.001; F(4,104) = 153.01, p < 0.001;) and over lateral more than central sites (Laterality: F(4,104) = 35.02, p < 0.001; F(4,104) = 22.12, p < 0.001; F(4,104) = 10.30, p < 0.001; F(4,104) = 4.54, p = 0.011). P300 amplitude was enhanced in the negative context as compared to the neutral one for the 600–700 ms latency window (Context: F(1,26) = 48.44, p < 0.001), particularly in posterior-central locations, as revealed by the interaction Context × Laterality (F(4,104) = 3.27, p = 0.046), and also for the 900–1000 ms latency window (F(1,26) = 4.63, p = 0.041). P300 amplitude was larger for women than men for the 700–800 ms latency window (Gender: F(1,26) = 8.38, p = 0.008), particularly in the right hemisphere (Gender × Laterality: F(4,104) = 3.76, p = 0.025). A trend in the same direction was observed for the 600–700 ms latency window (Gender: F(1,26) = 3.94, p = 0.058). (A) Grand average waveforms for all subjects in standard trials for neutral and ... Fig. 2. (A) Grand average waveforms for all subjects in standard trials for neutral and negative condition at four parietal electrodes. Notice that the affective content of pictures, although task-irrelevant, was processed, as reflected by the amplitude enhancement of P300 for the negative images. (B) Average waveforms for neutral context and in standard trials separately for men and women at four parietal electrodes. Notice that women showed larger P300 amplitude than men for processing emotionally neutral images. Figure options 3.2.2. Emotional modulation of involuntary attention Difference waves obtained by subtracting the ERPs elicited to standard trials from those elicited to novel ones revealed the neuroelectric activation underlying novelty processing, which was characterized by a well-defined N1-enhancement/MMN, a novelty-P3, and a reorienting negativity (RON; Fig. 3A). (A) Distraction potential (novel – standard difference waveforms) for all ... Fig. 3. (A) Distraction potential (novel – standard difference waveforms) for all participants at FCz for neutral and negative emotional conditions. Amplitude of the early phase of novelty-P3 was enhanced in the negative context condition. (B) Distraction potentials on FCz for men and women separately, for both neutral and negative conditions. Enhancement of the early phase of novelty-P3 in negative context was greater for women. (C) Scalp distribution maps of early novelty-P3 (200–290 ms.) in the negative context condition for women and men separately. Figure options No significant difference in mean amplitudes was found with ANOVA performed over the negative electrophysiological response combining an enhancement of auditory N1 and MMN. The ANOVA on novelty-P3 revealed a main effect of Phase (F(1,26) = 6.83, p = 0.015), Frontality (F(1,26) = 12.82, p = 0.001) and Laterality (F(4,108) = 61.53, p < 0.001), and the interactions Phase × Laterality (F(2,52) = 3.67, p = 0.034), Laterality × Frontality (F(2,52) = 16.73; p < 0.001) and Laterality × Frontality × Phase (F(2,52) = 23.36, p < 0.001), thus supporting the common frontal distribution of this potential; there were also larger amplitudes in the right hemisphere, particularly for its late phase, and larger amplitudes of the early phase at central sites. This pattern of results is in agreement with that obtained in previous studies ( Escera et al., 1998 and Escera et al., 2001). The amplitude of the early phase of novelty-P3 was larger for the negative context (Context × Phase: F(1,26) = 5.17, p = 0.031). The general ANOVA on novelty-P3 also yielded some effects of the intersubject variable Gender. A multiple interaction of location, emotional context and gender (Gender × Frontality × Phase × Context: F(1,26) = 5.25, p = 0.030) was found. Further analysis for groups separately revealed significantly larger amplitudes of the early phase of novelty-P3 in women (Context × Phase: F(1,13) = 9.85, p = 0.008), which were not found in men ( Fig. 3B). The general ANOVA revealed an interaction of Laterality, Phase and Gender (F(2,52) = 3.47, p = 0.040). The early phase of novelty-P3 was shown to be significantly larger for women than for men (Laterality × Gender: F(2,52) = 3.85, p = 0.029; Gender: F(1,26) = 4.27, p = 0.048; Fig. 4B), as revealed by subsequent analyses for phases separately. (A) Distraction potential (novel – standard difference waveforms) for both ... Fig. 4. (A) Distraction potential (novel – standard difference waveforms) for both conditions at FCz for men and women. Amplitude of the early phase of novelty-P3 was larger for women. (B) Distraction potentials on FCz for neutral and negative conditions separately, for both men and women. Larger amplitudes of early-P3 for women in both conditions were found, but the difference was greater for the negative condition. (C) Scalp distribution maps of early novelty-P3 (200–290 ms.) in the neutral context condition for women and men separately. Figure options As for the RON component, a general three-factor ANOVA revealed a significant interaction yielding larger amplitudes for the negative context (Fig. 5) (Context: F(1,26) = 5.30, p = 0.030). A main effect of Frontality (F(2,52) = 11.56, p = 0.001) and Laterality (F(2,52) = 4.49, p = 0.024) was also shown, while a marginal effect was found for Gender (Gender: F(1,26) = 3.61, p = 0.069), revealing an attenuated RON for women as compared to men. (A) Distraction potential (novel – standard difference waveforms) for all ... Fig. 5. (A) Distraction potential (novel – standard difference waveforms) for all participants at four frontocentral electrodes for neutral and negative emotional conditions, showing a positive deflection on neutral condition, instead of the component RON. (B) Scalp distribution maps of RON (440–510 ms.) for all participants separately for neutral and negative context conditions. The map shows a centro-parietally distributed positive potential for neutral condition.