Impulsivity in juvenile delinquency: differences among early-onset, late-onset, and non-offenders.

INTRODUCTION

Involvement in antisocial and delinquent behaviour is common during adolescence (Carroll, 1995; Carroll et al., 2001; Houghton and Carroll, 2002, 2004; Moffitt, 1993; Oyserman and Saltz, 1993; Rigby et al., 1989). Youths account for a disproportionately large amount of offences relative to their numbers in the general population (Australian Institute of Criminology, 2004;Oyserman and Saltz, 1993). For some time, researchers have suggested that delinquent behaviour is a result of deficits in impulse control (Barratt and Patton, 1983; Eysenck and Eysenck, 1977; Robbins and Bryan, 2004; Romero et al., 2001; White et al., 1994). Moreover, the age of onset of antisocial behaviour has been identified as being of central importance in understanding future antisocial behaviour. Specifically, the earlier the onset of antisocial behaviour the more likely those individuals will engage in further delinquent acts (Piquero and Chung, 2001). Moffitt (1993) distinguished between two patterns of delinquency, as a function of time of onset. She proposed that late-onset, adolescent-limited (AL) antisocial behaviour emerges during puberty and is common, normative, and relatively temporary. Life-course-persistent (LCP) antisocial behaviour, although rare, originates early in life and is persistent and pathological.

Deficits in neuropsychological functioning are commonly associated with antisocial behaviour (Loeber, 1990; Moffitt, 1990). According to Moffitt (1993), the criminal behaviour of LCP offenders may be due to neuropsychological impairments sustained during childhood. Such deficits in neuropsychological functioning, such as self-control (especially impulse control), may serve to maintain antisocial behaviour throughout life. In contrast, Moffitt purports that antisocial behaviour that emerges during adolescence is, on this account, the result of an individual reaching biological maturity prior to reaching social maturity (where he or she has legal access to such liberties to consume alcohol, and operate a motor vehicle). Observing peers who already have the resources and privileges associated with social maturity, some individuals deal with the developmental tensions by turning to delinquency, illegitimate means of achieving rationally motivated social goals.

The distinction between LCP and AL patterns of offending offers important developmental insights into the nature of delinquent behaviour. Henry et al. (1996) showed that 50-60% of all crimes committed in the United States can be attributed to approximately 5% of offenders, most of whom have histories of early emerging problem behaviour, and are at risk of longer term criminal careers. The remaining 95% of offenders are classified in the AL subgroup; not only does their offending begin later in life, but it also occurs less frequently and tends to be less violent. Several studies in the U.S. and New Zealand have shown that LCP adolescents are characterized by a number of risk factors such as social and familial disadvantages, poor parenting, and impulsivity and attention deficits (Fergusson et al., 1996; Jeglum Bartusch et al., 1997;Moffitt and Lynam, 1994; Caspi et al., 1997; White et al., 1994).

There is extensive evidence that impulsivity, or the inability to regulate self-control, is an important determinant of delinquent behaviour (Farrington et al., 1990; Gottfredson and Hirschi, 1990; Kindlon et al., 1995; Vitacco and Rogers, 2001; White et al., 1994). Furthermore Vitacco et al. (2002) found that, among adjudicated male adolescents, scoring high on impulsivity predicted greater antisocial behaviour at an 18 month follow up. However, findings have been inconsistent and contradictory, partly because of definitional and measurement issues. The construct impulsivity does not appear to be used with one common definition, but covers several behaviours or inferred processes, including cognitive, behavioural, and personality variables. Impulsivity is seen as a specific construct of self-regulation covering a broad range of behaviours (Baumeister and Vohs, 2004; Eysenck and McGurk, 1980). Several deficits in various dimensions of impulsivity have been associated with delinquent behaviours, including, cognitive variables, such as the inability to exercise inhibitory control and the tendency to respond quickly without thinking due to differences in cognitive tempo; behavioural variables, such as willingness to trade accuracy for speed when processing information and risk-taking behaviours; and the personality variable of impulsiveness. Therefore, the present research aimed to incorporate these various dimensions of impulsivity through the selection of relevant measures based on cognitive, behavioural, and personality variables previously unaccounted for in studies of the impulsivity of juvenile delinquents.

Inhibitory control is associated with neuropsychological functioning and is concerned with impairments in the inhibition of a dominant response. Barkley (1997) suggested that inhibition be assessed by performance on cognitive and behavioural tasks that require withholding of responding, delayed responding, cessation of responding, and resisting distraction or disruption by competing events. In addition, deficits in executive functions have also been linked to impulsivity and delinquent behaviour. Specifically, Moffitt (1990) found that individuals who performed poorly on neuropsychological tests of frontal lobe functioning (executive functioning), such as the Stroop Colour and Word Test (SCWT), were more likely to be antisocial in nature.

Barratt and Patton (1983) have argued that the tendency to respond quickly without thinking arises from biologically determined differences in cognitive tempo. Lawrence and Stanford (1999) found that individuals with high impulsivity display lower accuracy and faster time estimation than individuals with low impulsivity. They argued that this provides the basis for impulsive behaviours such as making quick decisions and acting without thinking. Much earlier, researchers defined cognitive impulsivity as the speed and the number of errors associated with making a decision (Kagan et al., 1964). Dickman and Meyer (1988) hypothesised that impulsivity was associated with a preference for information-processing strategies that emphasise speed at the expense of accuracy. Dickman and Meyer's results suggest that in spite of their generally faster response times and higher error rates, highly impulsive participants carry out at least one stage of processing, response execution, just as slowly and accurately as other individuals. The authors suggested that the advantage of high versus low impulsive individuals will depend on the nature of the task itself, suggesting that a lack of sensitivity in the measures commonly used to assess individuals' bias toward speed or accuracy in information processing and the lack of empirical research for the speed-accuracy trade-off model of impulsivity was responsible for the mixed findings.

The relationship between risk-taking behaviours and delinquency has been clearly documented, with young offenders tending to take significantly more risks than non-offenders (Arnett, 1992; Luengo et al., 1994). Moreover, high impulsive adolescents and young adults have been found to more frequently engage in risk-taking behaviour than individuals with low levels of impulsivity (Stanford et al., 1996; Vitacco et al., 2002). Interestingly Daderman et al. (2001), using Zuckerman's sensation seeking model, investigated why some sensation seekers become well socialised and engage in socially acceptable thrills, while others become socially delinquent. They concluded that juvenile delinquents were not interested in socially desirable forms of sensation seeking, but in fact gained arousal of the mind and senses through novel situations such as social drinking, drug use, and gambling.

Personality traits are continuous dimensions that may be used as instruments for the analysis of individual differences in behaviour (Daderman, 1999). In investigating delinquency, Romero et al. (2001) suggested that particular attention has been paid to temperament (or personality) variables. The majority of researchers in this field have tended to adopt Eysenck's (1967) three fundamental …