کارکردهای اجرایی و علائم عمومی در رفتار ضد اجتماعی نوجوانان: مطالعه مقطعی بر روی یک نمونه ایتالیایی مجرمان دیررس

Abstract Executive cognitive functions (ECFs) and other cognitive impairments, such as lower IQ and verbal deficits, have been associated with the pattern of antisocial and delinquent behavior starting in childhood (early-onset), but not with late-onset antisocial behavior. Beyond objective measures of ECF, basic symptoms are prodromal, subjectively experienced cognitive, perceptual, affective, and social disturbances, associated with a range of psychiatric disorders, mainly with psychosis. The goal of the present study was to examine ECF and basic symptoms in a sample of late-onset juvenile delinquents. Two-hundred nine male adolescents (aged 15–20 years) characterized by a pattern of late-onset delinquent behavior with no antecedents of Conduct Disorder, were consecutively recruited from the Social Services of the Department of Juvenile Justice of the city of Messina (Italy), and compared with nonantisocial controls matched for age, educational level, and socio-demographic features on measures for ECF dysfunction and basic symptoms. Significant differences between late-onset offenders (completers = 147) and control group (n = 150) were found on ECF and basic symptoms measures. Chi-square analysis showed that a significantly greater number of late-onset offending participants scored in the clinical range on several ECF measures. Executive cognitive impairment, even subtle and subclinical, along with subjective symptoms of cognitive dysfunction (basic symptom), may be contributing factor in the development and persistence of antisocial behaviors displayed by late-onset adolescent delinquents. The findings also suggest the need for additional research aimed to assess a broader range of cognitive abilities and specific vulnerability and risk factors for late-onset adolescent offenders.

Introduction In the development of antisocial behavior, emotional disturbances are frequently associated with neuropsychological dysfunctions and disruptive behaviors. In juvenile offender samples, symptoms of frontal lobe dysfunctions, such as inability to plan, poor insight, inflexible thinking, attentional difficulties, and impulsivity, were found [1] and [2]. Executive cognitive functioning (ECF) is a complex construct involving several dissociable abilities, such as cognitive flexibility, attention, planning abilities, concept formation, set-shifting, selective attention, abstract reasoning, problem-solving, inhibitory control, and working memory [3]. Subclinical impairment in ECF, which is diagnosable by neuropsychological assessments, may derive from a variety of biological and environmental factors in the context of a biosocial model of vulnerability in which genetic influences interact with perinatal complications and psychosocial factors [4]. Beyond objective measures of executive cognitive functions, little attention has been paid to subjective cognitive impairment in antisocial adolescents. A well-known trend in research on psychosis focuses attention on subjective complaints described as Basic Symptoms (BS). Basic symptoms are subjectively experienced cognitive, perceptual, affective, and social disturbances often recognized by the affected person long time before the outbreak of full-blown psychosis [5]. Several dimensions of the basic symptoms can be regarded as the subjective expressions of a pre-existing neuropsychological deficit in schizophrenia; cognitive and perceptive dimensions, such as thought interference, thought pressure, or decreased ability to discriminate between ideas and perception, are considered the most sensitive and specific basic symptoms for transition to psychosis, and they are reported by the 65% of those subjects who progress from prodromes to the psychotic illness [6]. On the other hand, recent evidence showed that basic symptoms are prevalent in adolescent general population, as 30.2% have at least one [7], in patients affected by Bipolar I and Bipolar II disorders [8], and in euthymic bipolar patients [9]. Furthermore, it has been shown that, in adolescents, basic symptoms in association with personality traits present as a nonspecific indicator of psychopathology rather than as an indicator of vulnerability to schizophrenia [10]. It actually seems that basic symptoms should be viewed as infrequent experiences not commonly occurring in adolescence and, consequently, they might possess a broad psychopathological meaning [11]. From a developmental point of view, Moffitt [12] outlined a dual taxonomy consisting of two major offending patterns: late-onset, adolescence-limited (AL), and early-onset, life-course persistent (LCP) antisocial behaviors. The first group consists of adolescents whose late-onset antisocial behavior is thought to be transient, representing a pattern of rebellious behavior arisen as a result of the social mimicry of deviant peers and of poor parental monitoring. The Author considered this pattern restricted to adolescence, characterized by sporadic crime-free periods and recovery by early adulthood, and largely contingent on environmental influences; the prevalence of adolescent-limited trajectory is high, as it is thought to include 15% to 30% of the adolescent population [13]. With the progression of chronological age, the assumption of more legitimate adult roles induces adolescence-limited delinquents to desist from criminal behavior. Early-onset, life-course persistent (LCP) group includes youths with conduct problems whose developmental trajectories are characterized by the influence of multiple risk factors. It has been suggested that prenatal and perinatal disruptions in neural development lead to executive and verbal deficits, engendering differences between children in emotional reactivity, behavioral regulation, level of activity, self-control, motor coordination, and cognitive abilities [12] and [14]. The main distinctive features of these offenders include early intense aggressiveness, cognitive deficits and impulsivity, severe family adversity, childhood violence exposure, and social disadvantages [15]. The distinction between the two patterns of offending offered significant developmental insights into the nature of delinquent behavior: in particular, the poor long-term prognosis for early-onset disruptive behaviors has been repeatedly confirmed [16] and [17]. It has been widely accepted that early- and late-onset offence trajectories differ in a number neuropsychological and psychopathological features; however, in the most recent literature the evidence of marked differences between the two patterns is less consistent [18], since also late-onset offenders displayed enduring problems at follow-up [19], and some of the same risk factors as persistent offenders, notably a convicted parent, a disrupted family, low verbal and nonverbal IQ, and hyperactivity in childhood [20]. Based on this background, the aim of the present research was to compare late-onset offenders with non-offending youths on a range of measures pertaining to executive cognitive functioning and basic symptoms. In light of previous findings summarised above, the broad prediction was that late-onset offenders would differ from non-antisocial controls matched for age and educational level on neuropsychological tasks and on basic symptoms measures.

Results The mean age of late-onset delinquent adolescents was 17.07 years (SD = 1.1), compared with a mean age of 17.29 (SD = 0.9) for the controls (t = −1.722, p = 0.86). Educational level (years) was 10.32 (SD = 2.51) for juvenile delinquents and 10.89 (SD = 2.48) for control subjects (t = −.282, p = 0.202). Socioeconomic status (SES; determination of socioeconomic status was based on income data) was 980€ (SD = 428.21) for juvenile delinquents and 997.86 (SD = 377.70) for control subjects (t = .376, p = .122). Types of delinquency found in the sample of late-onset offenders were violent crimes including assault/physical injuries, rape, robbery (n = 43, 29.25%), property offences, such as extortion, vandalism, burglary, motor vehicle theft, arson (n = 78, 53.06%), and other miscellaneous offences (n = 26, 17.68%). The mean scores, the standard deviations, and between-group differences for Stroop task, AB–AC, SPM, Phonemic Fluency, Semantic Fluency, and WCST CA are given in Table 1. Student t test for two independent samples evidenced significant differences between late-onset offenders and control groups on the following tasks: Stroop (t = 4.741; p = <0.0001), Phonemic Fluency (t = −14.701; p = <0.0001), Semantic Fluency (t = −3.010; p = .003), SPM (t = −10.309; p = <0.0001), and executive functioning as evidenced by the index “categories achieved” (CA) of the WCST (t = −3.180; p = .002). To test for the potential role of IQ on the ability to perform on the neuropsychological tests, it was decided to carry out a univariate comparison for this variable. Among the different neuropsychological tasks, the Stroop test, F (1, 285) = .773, ns ceased to be a significant predictor of group differences. Table 1. Comparisons between mean scores on neuropsychological tasks in late-onset offenders (n = 147) and controls (n = 150). Scores have been compared by Student’s T Test for two independent samples. Late-onset offenders Controls Student’s T Test Mean S.D. Mean S.D. t p Stroop 41.79 29.87 28.03 18.65 4.741 <.0001 Phonemic Fluency 17.92 6.70 29.28 6.45 −14.701 <.0001 Semantic Fluency 32.69 9.19 36.03 9.68 −3.010 .003 SPM 94.77 16.36 112.18 12.14 −10.309 <.0001 AB–AC 1.49 4.57 1.36 2.33 .315 .753 WCST PE 17.53 13.17 15.06 10.48 1.772 .077 WCST CA 5.09 1.51 5.55 0.89 −3.180 .002 SPM = Standard Progressive Matrices; WCST PE = Wisconsin Card Sorting Test Perseverative Errors; WCST CA = Wisconsin Card Sorting Test Categories Achieved. Table options Late-onset offenders exhibited slower reaction times on the interference trial of the Stroop color–word set, although it should be noted that these differences were within the healthy functioning range. When required to inhibit a previous learned response pattern (conflict blocks), late-onset offenders displayed greater reaction time, though they were more likely to experience higher levels of interference than controls. On the Verbal Fluency test, late-onset offenders, compared with controls, produced fewer words on Phonemic Fluency task, and category names on Semantic Fluency task. On measures of intellectual functioning significant differences between the groups in full scale IQ were found; on SPM, the average for late-onset offenders was 35 items correct, which in the Italian standardization correspond to the 51st percentile with an IQ equivalent of 94.77, whereas the average for controls was 46 items correct, corresponding to the 59th percentile and to an IQ equivalent of 112.1. Late-onset offenders did not demonstrate worse performance than controls on tasks measuring proactive interference in associative memory (AB–AC) (t = .315; p = .753). Regarding the frequencies (expressed in percentages) of subjects who reported scores in the clinical range within the two groups (Table 2), 19 subjects (12.9%) of the late-onset offender group exhibited performance deficits on Stroop task, vs none of the community control group. The higher frequency of pathological subjects in the late-offender group was found on the Phonemic Fluency task: 123 subjects out of 147 (83.6% of the sample) vs only 22 controls (14.7%) showed an impairment of the rapid intrinsic response generation where responses were minimally specified by external cues or triggers; only minimal differences between groups were found on Semantic Fluency task, with 44 late-onset offenders vs 50 controls exhibiting an impaired performance. Regarding SPM, 17 late-onset offenders (11.4%) had IQ equivalent below the norm. Moreover, 42 (28.6%) late-onset offenders vs 24 (16%) controls did not complete a sufficient number of categories on the WCST. Neither subjects nor controls exhibited performance deficits on AB–AC task. Table 2. Frequencies of subjects with pathological scores on neuropsychological tasks in late-onset offenders (n = 147) and controls (n = 150). Late-onset offenders Controls Chi-square test n % n % χ2 p Stroop 19 12.9 0 0 20.562 <.0001 Phonemic Fluency 123 83.6 22 14.7 1.378 <.0001 Semantic Fluency 44 30.0 50 33.3 .371 .542 SPM 17 11.4 0 0 18.144 <.0001 AB–AC 0 0 0 0 – – WCST CA 42 28.6 24 16.0 6.654 0.01 SPM = Standard Progressive Matrices; WCST CA = Wisconsin Card Sorting Test Categories Achieved. Table options Regarding basic symptoms, the means and the standard deviations for the self-rating scale FBF are shown in Table 3. Late-onset offenders had a much higher FBF total score than normal controls (t = 4.950; p = > .0001), and FBF subscales’ scores were also much higher than in controls. As shown by Student t test for two independent samples, late-onset offenders had a significantly higher rate of basic symptoms than controls on the following FBF domains: loss of control (KO; p = <.0001), language (SP; p = <.0001), thought (DE; p = .007), memory (GED; p = .047), motility (MO; p = .002), lack of automatism (AU; p = <.0001), anhedonia and anxiety (AN; p = <.0001), and sensorial overstimulation/overinclusion (REI; p = <.0001). Table 3. Comparisons between mean scores on clinical scales in late-onset offenders (n = 147) and controls (n = 150). Scores have been compared by Student’s T Test for two independent samples. Late-onset offenders Controls Student’s T Test Mean S.D. Mean S.D. t p FBF Loss of Control 2.44 1.57 1.53 1.408 5.230 <0.0001 Simple perception 1.31 1.89 1.11 1.301 1.059 .291 Complex perception 1.23 1.64 0.93 1.369 1.670 .096 Language 2.61 2.25 0.96 1.404 7.566 <0.0001 Thought 2.55 2.14 1.88 2.072 2.710 .007 Memory 1.95 2.09 1.51 1.694 1.993 .047 Motility 1.86 1.79 1.25 1.572 3.053 .002 Lack of automatism 2.22 2.12 1.38 1.557 3.871 <0.0001 Anhedonia/Anxiety 1.46 1.68 0.55 1.145 5.432 <0.0001 Sensorial overstimulation/Overinclusion 2.61 2.16 1.73 1.968 3.600 <0.0001 Total score 20.21 13.88 12.83 11.47 4.950 <0.0001 Table options To test for the potential role of depressive symptoms or anxiety – which usually develops after incarceration – to perform on the neuropsychological tests, it was decided to carry out a univariate comparison for the anhedonia and anxiety variable of the basic symptoms (AN). Again, among the different neuropsychological tasks, the Stroop test, F (1, 285) = .484, ns ceased to be a significant predictor of group differences. The frequencies (expressed in percentages) of subjects who reported higher scores within the two groups are shown in Table 4. In the late-offenders group, loss of control (n = 42; 28.6%), sensorial overstimulation/overinclusion (n = 31; 21.4%), thought (n = 28; 19.3%), and language (n = 26; 17.9%) were the most frequent basic symptoms. Table 4. Frequencies of subjects with pathological scores on FBF in late-onset offenders (n = 147) and controls (n = 150). Late-onset offenders Controls Chi-square test n % n % χ2 p FBF Loss of Control 42 28.6 9 6 26.274 <0.0001 Simple perception 8 5.7 0 0 8.815 .003 Complex perception 4 2.9 0 0 4.346 .037 Language 26 17.9 4 2.7 18.566 <0.0001 Thought 28 19.3 12 8 7.924 .005 Memory 12 8.6 4 2.7 4.843 .028 Motility 8 5.7 8 5.3 .020 .887 Lack of automatism 20 13.6 4 2.7 11.793 0.001 Anhedonia/Anxiety 6 4.3 0 0 6.564 0.010 Sensorial overstimulation/Overinclusion 31 21.4 8 5.3 16.475 <0.0001