روابط افتراقی تکانشگری و یا نشانه های ضد اجتماعی در P50، N100، یا P200 راهگاهی حسی شنوایی در گروه شاهد و اختلال شخصیت ضد اجتماعی

Abstract Limited information is available on the relationship between antisocial personality disorder (ASPD) and early filtering, or gating, of information, even though this could contribute to the repeatedly reported impairment in ASPD of higher-order information processing. In order to investigate early filtering in ASPD, we compared electrophysiological measures of auditory sensory gating assessed by the paired-click paradigm in males with ASPD (n = 37) to healthy controls (n = 28). Stimulus encoding was measured by P50, N100, and P200 auditory evoked potentials; auditory sensory gating (ASG) was measured by a reduction in amplitude of evoked potentials following click repetition. Effects were studied of co-existing past alcohol or drug use disorders, ASPD symptom counts, and trait impulsivity. Controls and ASPD did not differ in P50, N100, or P200 amplitude or ASG. Past alcohol or drug use disorders had no effect. In controls, impulsivity related to improved P50 and P200 gating. In ASPD, P50 or N100 gating was impaired with more symptoms or increased impulsivity, respectively, suggesting impaired early filtering of irrelevant information. In controls the relationship between P50 and P200 gating and impulsivity was reversed, suggesting better gating with higher impulsivity scores. This could reflect different roles of ASG in behavioral regulation in controls versus ASPD.

. Introduction Antisocial Personality Disorder (ASPD) is a serious pathology associated with changes in evoked potential components reflecting impaired higher-order information processing (Bauer, 2001, Chang et al., 2010 and Gao and Raine, 2009). Considerably less is known about relationships between ASPD and early stimulus encoding or filtering, although these processes may influence later information processing affected in ASPD (Boutros et al., 2004 and Gjini et al., 2010). We studied pre- and early-attentional information processing in subjects with ASPD, and investigated relations with symptom severity and impulsivity, a key feature of ASPD (Swann et al., 2009). Early information processing can be studied with the paired-click paradigm, a passive listening task in which two identical click stimuli are presented in rapid succession. The first click (S1) elicits P50, N100, and P200 auditory evoked potentials, reflecting stimulus encoding. The second click (S2) elicits corresponding, but attenuated, potentials (Fruhstorfer et al., 1970), reflecting filtering (auditory sensory gating, ASG) of information. ASG may reflect inhibitory mechanisms protecting higher-order functions from irrelevant information (Freedman et al., 1991). P50, N100, and P200 components reflect different underlying mechanisms and functions (Boutros et al., 2004, Brockhaus-Dumke et al., 2008, Crowley and Colrain, 2004 and Jansen et al., 2004): P50 appears to be related to pre-attentional processes (Näätänen, 1992), N100 to early triggering of attention (Näätänen, 1992 and Rinne et al., 2006), and P200 to early allocation of attention and initial conscious awareness (Näätänen, 1992). Preliminary results showed later P50 peak latencies and reduced P50 ASG in subjects with ASPD, but not in subjects with adult-onset antisocial behavior, compared to controls (Lijffijt et al., 2009). There was also a trend for a more pronounced P50 ASG impairment in subjects endorsing more conduct disorder symptoms, similar to relationships between ASPD symptoms and changes in evoked potentials reflecting higher-order mechanisms (Bauer, 2001 and Chang et al., 2010). These outcomes suggest delayed pre-attentional stimulus encoding and impaired pre-attentional filtering in ASPD. However, study samples were small, and impaired filtering could have been moderated by co-occurring past alcohol or drug use disorders, which frequently coexist in subjects with ASPD (Goldstein et al., 2006, Goldstein et al., 2007, Krueger et al., 2002 and Krueger et al., 2005), and are also potentially related to impaired P50 gating (Boutros et al., 2006, Fein et al., 1996, Fuentemilla et al., 2009, Marco et al., 2005, Patrick et al., 1999, Rentzsch et al., 2007 and Thoma et al., 2006), although not all studies showed this (Adler et al., 2001, Boutros et al., 2000a, Boutros et al., 2000b and Fein et al., 1996). N100 and P200 ASG have not been studied in ASPD, although subjects with substance use disorders had reduced N100 or P200 gating compared to healthy controls (Boutros et al., 2000a, Boutros et al., 2006 and Fuentemilla et al., 2009). These results suggest impaired P50, N100, or P200 gating in subjects with substance use disorders, potentially moderating gating deficits in ASPD. This would be consistent with relationships between ASPD or antisocial traits and enhanced early-attentional orienting to stimuli as measured by an increase in N100 or N100-like components, which has been interpreted as increased processing of potentially irrelevant information (Franken et al., 2005, Houston and Stanford, 2001 and Liu et al., 2007). Automatic orientation to stimuli seems to be stronger in subjects who score higher on impulsivity (Franken et al., 2005 and Hegerl et al., 1995), which is enhanced in subjects with ASPD (Swann et al., 2009). These results suggest ASPD could be related to increased automatic early-attentional triggering or orientation, potentially resulting in a smaller difference in N100 or P200 amplitude between S1 and S2, reflecting impaired sensory gating. Impaired gating in ASPD might be moderated by impulsivity, a predisposition to action without planning or regard for consequences (Moeller et al., 2001) that is prominent in ASPD and antisocial behavior in general (Cale, 2006, Luengo et al., 1994 and Swann et al., 2009), as well as in substance use disorders (Krueger et al., 2002, Moeller et al., 2001 and Ruiz et al., 2008). Although substance use disorders and impulsivity overlap they could exert different effects on information processing in ASPD (Swann et al., 2009). We investigated P50, N100, and P200 ASG in males with ASPD, expecting impaired P50, N100, and P200 ASG in ASPD compared to healthy controls. Deficits could be related to: 1) ASPD per se, 2) co-occurring substance use disorders, or 3) impulsivity. If impaired ASG relates to ASPD per se we would expect gating deficits irrespective of co-occurring substance use disorders, with a potential relationship between ASG and ASPD symptom count. If deficits are related to substance use disorders, we would expect impairments in subjects with combined ASPD and substance use disorders, but not ASPD only. Finally, if impaired gating is related to increased impulsivity we would expect more pronounced deficits in subjects with higher trait impulsivity. Effects of substance use disorders were investigated in subjects with ASPD with past substance use disorders because P50, N100, and P200 gating might improve during abstinence (Boutros et al., 2006), thus reducing possible confounding acute effects of substances.

. Results 3.1. Group differences Fig. 1 shows P50, N100, and P200 evoked potentials. Table 2 shows outcomes for electrophysiology and statistical tests. P50, N100, and P200 amplitudes were smaller, and N100 and P200 peaked earlier, for S2 than for S1. Group effects were not significant (ESs < 0.5, 95%-CIs including 0). There were no effects of smoking. GLM analyses with group and history of HI revealed no differences in evoked potential or gating measures between subjects with (n = 17) or without a history of head injury (n = 47) (F < 3.70, p > .06). This suggests history of HI with no residual after-effects did not relate to changes in sensory processing or gating. Averages for controls and subjects with ASPD for P50 potentials (left panel, ... Fig. 1. Averages for controls and subjects with ASPD for P50 potentials (left panel, filter 10–50 Hz), and N100 and P200 potentials (right panel, filter 1–20 Hz). Figure options Table 2. Electrophysiology for controls (NC, n = 28) and subjects with antisocial personality disorder (ASPD, n = 37). Group Statistics NC ASPD Stimulus (S)a Group (G) S∗G P50 componentb Latency S1 60.96 ± 12.24 60.33 ± 14.86 Z = −0.71 N/A Z = −0.10 S2 57.75 ± 10.71 59.54 ± 15.65 Z = −0.39 Amplitude S1 2.51 ± 1.39 2.06 ± 1.16 F (1,62) = 57.71 F (1,62) = 0.29 F (1,62) = 1.53 S2 1.24 ± 1.07 1.24 ± 1.0 Ratio 55.22 ± 46.50 63.53 ± 45.53 N/A F (1,62) = 0.68 N/A Difference score 1.27 ± 1.38 0.90 ± 1.08 N/A F (1,62) = 1.41 N/A N100 component Latency S1 104.82 ± 11.74 109.95 ± 15.15 Z = −4.71 N/A Z = −1.35 S2 93.43 ± 11.07 98.39 ± 14.92 Z = −1.18 Amplitude S1 −6.54 ± 3.72 −5.82 ± 3.38 F (1,63) = 112.27 F (1,63) = 0.01 F (1,63) = 3.08 S2 −3.05 ± 1.53 −3.55 ± 2.08 Ratio 53.32 ± 27.89 64.23 ± 27.0 N/A F (1,63) = 2.69 N/A Difference score −3.50 ± 2.94 −2.27 ± 2.21 N/A Z = −1.82 N/A P200 component Latency S1 205.18 ± 27.02 199.95 ± 27.16 F (1,63) = 22.37 F (1,63) = 0.12 F (1,63) = 0.82 S2 184.89 ± 27.86 186.16 ± 26.47 Amplitude S1 10.75 ± 5.41 9.62 ± 5.53 F (1,63) = 196.14 F (1,63) = 0.31 F (1,63) = 2.27 S2 4.86 ± 2.09 4.97 ± 2.30 Ratio 51.84 ± 25.84 56.16 ± 17.04 N/A F (1,63) = 1.12 N/A Difference score 5.89 ± 4.22 4.66 ± 3.93 N/A Z = −1.43 N/A Bold: p < .05. a Difference testing between S1 and S2 across groups. b ASPD: n = 36 for P50 S2, ratio, and difference score due to n = 1 with P50 S1 = 0 uV. Table options 3.2. Effects of past substance use disorder Subjects had no substance use disorder (28 controls, 7 ASPD), past alcohol use disorder (2 subjects), past drug use disorder (7 subjects), and past alcohol + drug use disorder (21 subjects). Subjects with past substance use disorder were in early partial (6 subjects), early full (6 subjects), or sustained full remission (18 subjects). GLM revealed no significant effects of past substance use disorder. Dunnett tests, comparing controls with ASPD without substance use disorder or with combined alcohol + drug use disorder, revealed outcomes comparable to those obtained with GLM analysis. Duration of abstinence was not tested further due to small samples for early partial and early full remission. 3.3. Effects of symptom count Table 3 shows correlations between ASPD symptom counts and P50 variables for ASPD. Symptom counts correlated significantly with P50 ratio and difference score. Fig. 2 depicts the relationship between ASPD total symptoms and P50 ratio. No significant correlations were found for N100 or P200 variables. Table 3. Antisocial personality disorder (ASPD) symptom counts, BIS-11 impulsivity scores, and correlations with P50 S1 and S2 amplitudes, P50 S2/S1 ratio, and S1 − S2 difference score (P50 Δ). Group Statistics ES Correlations (r) NC ASPD P50 ratio P50 Δ P50 S1 P50 S2 Symptom count (median [range]) ASPDb CD 0 (0–1) 5 (2–14) 0.33 −0.42 −0.24 0.19 AAB 0 (0–1) 4 (3–6) 0.24 −0.24 −0.12 0.21 ASPD 0 (0–2) 9 (5–20) 0.41 −0.48 −0.24 0.24 BIS-11 (mean [SD])a NC/ASPDb Non-planning 21.3 (4.5) 24.9 (4.7) F(1,63) = 9.8† −0.79 −0.32/0.29† 0.23/−0.10 −0.03/0.08 −0.36/0.31† Motor 20.6 (3.6) 24.2 (4.0) F(1,63) = 14.5† −0.95 −0.37/0.24† 0.23/−0.24† 0.11/−0.16 −0.21/0.09 Attentional 13.4 (3.3) 16.2 (3.9) F(1,63) = 10.0† −0.79 −0.55/0.31† 0.46/−0.23† 0.12/0.01 −0.44/0.26† Total 55.3 (7.4) 65.3 (10.0) F(1,63) = 19.5† −1.11 −0.55/0.35† 0.41/−0.23† 0.09/−0.02 −0.46/0.28† Italics: Kendall’s tau instead of Pearson r, Underlined: p < .05. †Between-group difference p < .05. a Statistical outcomes based on logarithmically transformed data. b ASPD: n = 36 for P50 S2 amplitude, ratio, and difference score. Table options Relationship between ASPD total symptom count and P50 ratio (r = 0.41) across ... Fig. 2. Relationship between ASPD total symptom count and P50 ratio (r = 0.41) across subjects with ASPD. Figure options 3.4. Effects of impulsivity 3.4.1. BIS-11 and P50 Table 3 shows group differences for BIS-11, and correlations between BIS-11 and P50 variables. Subjects with ASPD had significantly higher BIS-11 scores than controls. BIS-11 total score correlated positively with P50 ratio in subjects with ASPD, but negatively with P50 ratio in controls (Table 3). BIS-11 total score also correlated with P50 difference score and P50 S2 amplitude in controls only. These correlations differed significantly between groups (r-to-z tests, z > |2.55|, p < .011). Fig. 3 suggests a non-linear relationship between impulsivity and P50 gating across all subjects. Non-linear regression across all subjects revealed a significant quadratic relationship between BIS-11 total score and P50 ratio (F(1,61) = 3.70, p = .031), and a marginally significant quadratic relationship with P50 difference score (F(1,61) = 3.09, p = .053). Non-linear regression analyses for each group separately revealed a significant quadratic relationship between BIS-11 total score and P50 ratio only in controls, which seemed less pronounced compared to the linear relationship (F = 11.46 and 5.81 for linear and quadratic relationships, respectively). These outcomes cannot distinguish between a linear, but opposite, relationship between impulsivity and P50 gating in controls and subjects with ASPD, or a quadratic relationship with controls and subjects with ASPD falling differently along the curve. Relationship between BIS-11 total score and P50 ratio across controls (upper ... Fig. 3. Relationship between BIS-11 total score and P50 ratio across controls (upper panel, r = −0.55) and subjects with ASPD (lower panel, r = 0.35). Figure options 3.4.2. BIS-11 and N100, P200 In ASPD, BIS-11 total, motor, and attentional scores correlated significantly with N100 difference score (tau = 0.26, 0.23, 0.24, p < .05), although these correlations may not differ significantly from those found in controls (tau = 0.13, 0.05, 0.17). As N100 amplitudes and difference scores were expressed as negative values, positive correlations suggest worse N100 gating with increased impulsivity. In controls, BIS-11 attentional score correlated with P200 difference score (r = 0.40, p = .034; ASPD r = −0.08), suggesting better P200 gating with increased impulsivity. No significant relationships were found with S1 or S2 amplitudes. 3.4.3. BIS-11 and latencies In controls, P50 S2 latency correlated with attentional score (tau = 0.35, p = .027; ASPD tau = 0.10), N100 S2 correlated with non-planning score (tau = −0.33, p = .018; ASPD tau = 0.13), and P200 S2 correlated with motor score (tau = −0.30, p = .029; ASPD tau = −0.03). As correlations may differ between ASPD and controls for the relationship between BIS-11 and P50, N100, and P200 S2 latency, outcomes suggest increased impulsivity could relate to later P50, but earlier N100 and P200 S2 peaking in controls. 3.5. Gating in ASPD: symptoms versus impulsivity In ASPD, outcomes for P50 ASG showed significant relationships with symptom count and impulsivity. ASPD total symptoms correlated significantly with BIS-11 total score (r = 0.42, p = .009). To disentangle effects of symptoms and impulsivity, we conducted GLM analyses with P50 ratio and difference score as dependent variables, BIS-11 total score and ASPD total symptom count as continuous variables, and past substance use disorders as dichotomous variables. Total symptom count, but not BIS-11 total score, related to P50 difference score (F(1,29) = 7.25, p = .012) and, less pronounced, to P50 ratio (F(1,29) = 3.36, p = .077). This suggests that in ASPD P50 gating is impaired in subjects endorsing more symptoms.