Table 2.
Study | N | Type & age | % ♀ | Exclusion criteria or covariates | Behavioral measures | Method | Key findings |
---|---|---|---|---|---|---|---|
Breeden et al., 2015 | 47 | 1 M (SD) = 14 (3) |
47 | Exclusions: IQ < 80 Covariates: age &IQ |
AB: SDQ & CBCL (P) Psychopathic/CU traits: ICU (PC) |
ROI-A: (TBSS) |
Dimensional: CU traits: ↓ FA bilateral UF, & right fornix, controlling for overlapping externalizing and similar when examining only within AB + group AB: No associations found controlling for CU traits Categorical: AB + CU- vs. HC: ↓ FA right UF, right fornix; AB + CU + vs. HC: ↓ FA bilateral UF, bilateral fornix; AB + CU + vs. AB + CU −: no significant differences |
Decety et al., 2015 | 110 | 1 & 2 M (SD) = 10 (1) |
52 | Exclusions: developmental disorder, head trauma Covariates: age, gender & race (also explored as moderators) |
AB: DISC (P&S) | Whole brain (TBSS) |
Dimensional: CD symptoms: ↑ AD in left CT, right SCR, & right Fmin but ↑ RD in right SLF – findings particularly pronounced for females & African–American participants relative to Caucasian/Hispanic. Categorical: AB + vs. subAB: no significant differences subAB + vs. HC: no significant differences AB + vs. HC: no significant differences but gender & race interactions (AD & RD) - regions comparable to above |
Finger et al., 2012 | 31 | 1 & 3 M (SD) = 14 (2) |
10 | Exclusions: psychiatric or neurological disorder, head trauma, IQ < 80 Covariates: age, IQ |
AB: KSADS DSM-IV (I) Psychopathic/CU traits: ASPD (P) & PCL:YV (I) |
Whole brain (TBSS) & ROI-A: (DT) | Categorical: AB + P + vs. HC: no significant differences |
Haney-Caron et al., 2014 | 41 | 1 & 2 M (SD) = 16 (1) |
39 | Exclusions: any other psychiatric disorder Covariates: gender, CD age of onset, & IQ |
AB: KSADS DSM IV (I) | Whole brain: (TBSS) |
Dimensional: CD symptoms: ↓ FA right ACR, bilateral SCR, IFOF, ILF, & left CC & PCR. ↑ FA bilateral ATR (similar controlling for CD onset age, gender, & IQ) Categorical: AB + vs. HC: ↓ FA bilateral SCR, ACR, IFOF, SLF, ILF, IC, & ped. ↓ AD bilateral SCR, ACR, IFOF, SLF, ILF, IC, ped, UF, CC, CG, Fmin, & right PCR |
Hummer et al., 2015 | 66 | 1 M (SD) = 14 (1) |
27 | Covariates: Age (also examined as moderator) | AB: KSADS DSM-IV (I) | ROI-A: (TBSS) | Categorical: AB + ADHD − vs. HC: no significant differences AB + ADHD + vs. HC: no significant differences AB + vs. HC interaction with age: ↑ FA with increasing age only in HC vs. AB groups in CC & left SLF |
Li et al., 2005 | 76 | 1 M (SD) = 14 (1) |
39 | Exclusions: HC: no psychiatric disorder Covariates: age |
AB: KSADS DSM-IV (I) | Whole brain: (VBA) | Categorical: AB + vs. HC: ↓ FA bilateral SLF |
Pape et al., 2015 | 145 | 4 M (SD) = 18 (2) |
14 | Covariates: age & IQ | Psychopathic/CU traits: YPI (S) | Whole brain (TBSS) | Dimensional: CU traits: ↑ AD bilateral CT Grandiose-manipulative traits: ↑ FA and ↓ RD bilateral ATR, CT, Fmin, IFOF, UF, CC; ↓ RD bilateral CG, Fmaj, ILF, SLF |
Passamonti et al., 2012 | 26 | 1 & 2 M (SD) = 18 (1) |
0 | Exclusions: developmental disorder, physical illness, IQ < 85; HC with IQ > 115 Covariates: ADHD |
AB: KSADS DSM IV (I) Psychopathic/CU traits: YPI (S) |
Whole brain (VBA) & ROI-M (DT) | Categorical: AB + vs. HC: ↑ FA, ↑ AD, & ↓ RD bilateral external capsule; (whole-brain); ↑ FA, ↑ AD, & ↓ RD bilateral UF (ROI). Similar accounting for effects of ADHD symptoms & hemispheres. |
Peper et al., 2015 | 258 | 1 M (SD) = 14 (3) |
52 | Exclusions: neurological, psychiatric or endocrine illness. Covariates: age |
AB: Buss Perry Aggression (S) | ROI-A (DT) |
Dimensional: Age: ↑ FA overall Verbal aggression: ↑ MD temporal–temporal Hostility: ↓ MD in subcortical-temporal & temporal-temporal |
Sarkar et al., 2013 | 43 | 1 & 2 M (SD) = 16 (2) |
0 | Exclusions: IQ < 80 Covariates: age |
AB: KSADS DSM-IV (I), SDQ (P&S) Psychopathic/CU traits: PCL:YV (I) &APSD (P&S) |
ROI-M: (DT) |
Dimensional: SDQ (total & conduct problem scales): ↑ FA bilateral UF; APSD (total & CU traits scales): ↑ FA bilateral UF Categorical: AB + P + vs. HC: ↑ FA left UF |
Zhang et al., 2014b | 69 | 1 & 3 M (SD) = 15 (1) |
0 | Exclusions: psychiatric disorder, head trauma, past year substance abuse, IQ ≤ 80 | AB: SCID DSM-IV (I), SDQ & BIS (S) Psychopathic/CU traits: APSD (S) |
Whole brain (TBSS) & ROI-A (DT) |
Dimensional: no significant associations Categorical: AB + vs. HC: ↑ FA bilateral CC, left ACR, right SCR; ↓ RD bilateral CC, left ACR |
Zhang et al., 2014a | 56 | 1 & 3 M (SD) = 14 (1) |
46 | Exclusions: psychiatric disorder, substance use, head trauma, IQ ≤ 80 Covariates: gender tested as moderator |
AB: SCID DSM-IV (I) & SDQ (S) CU traits: APSD (S) |
ROI-A: (VBA & DT) |
Dimensional: CU traits: ↓ RD & ↓ MD right UF (boys with CD only; uncorrected) Categorical: AB + vs. HC: no significant differences |
Note that we report findings from the most stringent analyses carried out by included studies, including models controlling for multiple comparisons or overlap of outcome variables.
Type of sample: 1 = community; 2 = high-risk; 3 = clinic; 4 = forensic.
DTI acronyms: ACR = anterior corona radiata; AD = axial diffusivity; ATR = anterior thalamic radiation; CC = corpus callosum; CG = cingulum/cingulate gyrus; CT = corticospinal tract; Fmaj = forceps major; Fmin = forceps minor; FA = fractional anisotropy; IC = internal capsule; IFOF = inferior fronto-occipital fasciculus; ILF = inferior longitudinal fasciculus; PTR = posterior thalamic radiation; RD = radial diffusivity; SCR = superior corona radiata; SLF = superior longitudinal fasciculus; UF = uncinate fasciculus.
Behavior acronyms: AB = antisocial behavior; CU = callous-unemotional; P = psychopathy; S = self-reported; P = parent-reported; I = structured interview.
Behavior groupings: AB + P + = high on AB and psychopathy; AB + P- = high on AB and low on psychopathy; AB + CU + = high on AB and callous-unemotional traits; HC = healthy controls.
DTI method acronyms: DT = deterministic tractography; ROI-A = region of interest automatic generation; ROI-M = region of interest, manually drawn; TBSS = tract-based spatial statistics; VBA = voxel based analyses.