Adaptive decision making and patterns of risk orientation to potential gains and losses in emotionally and behaviourally disturbed adolescents: Implications for the reduction of dangerous risky-behaviour

Objectives: Adolescent risk-taking behaviour was reviewed from the perspective of decision-making theory. Patterns of risk-orientation for potential gains and losses, and adaptive decision making amongst adolescents with, and without, emotional and behavioural disturbance were predicted. These were tested in two computer-based tasks. Design & Methods: The design used was quasi-experimental. 130 early adolescents took part (internalizing, N=13, externalizing, N=11, and control sample, N=106). Participants were matched according to age, sex and score on Raven's Progressive Matrices. Participants completed the Multidimensional Measure of Children's Perception of Control, the Need for Closure Scale and an abbreviated version of Schneider & Lopes' (1986) Lottery Selection Task, a measure of risk seeking and risk aversion to potential losses and gains. Participants also played a computer game designed by the author which tests adaptive decision-making where both information and intervention have systematically varying costs. Results: Internalizing adolescents were found to have higher external locus-of-control than other groups. Group differences in risk-orientation to potential losses and gains were observed. Internalizers showed risk seeking for potential gains and risk aversion to potential losses. Externalizers showed the reverse pattern. The control group showed intermediate behaviour. External locus-of-control was predictive of risk-orientation to gains but not losses. There was evidence that higher ability adolescents make more use of information and that, convergent with their risk-orientation patterns, externalizers are less prepared to endure a cost to gain information. Need for closure measures showed no effects. Conclusions: That externalizers should be risk averse and internalizers should be risk-seeking for potential gains appears counter-intuitive. The results are understood in terms of Lopes' (1993) two-factor model of decision-making and highlight the importance of framing effects in preventative work with, and treatment of, adolescents exhibiting risky behavior