For best experience please turn on javascript and use a modern browser!
You are using a browser that is no longer supported by Microsoft. Please upgrade your browser. The site may not present itself correctly if you continue browsing.

Laboratory and real world observations of risk-taking behaviour among children and adolescents show conflicting results. In the real world adolescents take more risks than children, but in lab settings they engage in equal levels of risks. Ivy Defoe is developing a new psychological model that can explain these differences. ‘In the real world age and culture predict exposure to risk conducive situations, both offline and online. Increases in such risk exposures as youth become older, predict increases in youth risk behaviour. Risk exposure is thus important to explain age differences in risk-taking, but was so far lacking in contemporary psychological models on youth risk-taking development.’

young person is binge drinking

‘Laboratory experimental studies, that are regarded as a golden standard  for the scientific investigation of cause and effect, do not reflect the same pattern of age differences in risk-taking that we observe in the real-world’, states Ivy Defoe, assistant professor within the Forensic Child and Youth Care Sciences research group at the University of Amsterdam. Defoe based this conclusion on a systematic review (meta-analysis) published in Psychological Bulletin. As the lead author she summarized 21 laboratory studies comparing children (5-10 years) versus adolescent (11-19 years) risk-taking.

In the real-world adolescents are known to engage in heightened harmful risk-taking compared to other age groups, ‘think of binge-drinking, vandalism, or careless behavior in traffic’. But in the meta-analysis, Defoe and colleagues found that adolescents and children engage in equal levels of risks in laboratory settings. When adolescents are given the opportunity to reject a risky option and choose a sure ‘non-risky’ option instead, they even engage in fewer risks than children. Furthermore younger adolescents (11-13 years) take more risks than older adolescents (14-19) in the lab, whereas in the real-world, again the opposite is observed.

A new model to explain these differences

To explain these differences in risk-taking in lab settings versus the real world Defoe and colleagues introduced the Developmental Neuro-Ecological Risk-Taking Model (DNERM). This model predicts that adolescents engage in more risks than children, and older adolescents engage in more risks than younger adolescents in the real-world because as youth grow older they face more risk exposures.

‘Think of youth increasingly having access to alcohol when they grow older because the law allows them to, or to online platforms where they can commit fraud. Children and younger adolescents are less exposed to these type of risks in real life, and hence engage in fewer risk behaviors. In the controlled setting of the lab, risk exposure is equal for everyone, independent of age, leading to more equal levels of risks for children and adolescents, and to younger adolescents even engaging in more risks than older adolescents,’ explains Defoe.

As youth grow older they also increasingly come in contact with peers who are engaging in risk behaviour. Children and young adolescents do not typically have those opportunities of physical and social risk exposures. ‘Thankfully so, because, the findings of the meta-analysis suggest that if children and younger adolescents did have access to these “risk exposures”, they would engage in equal or even more risk behaviors than (older) adolescents’, states Defoe. Although virtually absent from psychology models on youth risk-taking development, this ecological concept of risk exposure is acknowledged to some extent in criminology models. ‘But both psychology and criminology models overlook the above-described age-dependent component of risk exposure.’

Self-control and culture

Besides age Defoe’s model also takes the role of self-control into account, illustrated by the case when the legal drinking age switched from 16 to 18 in the Netherlands in 2014: ‘When there is access to alcohol, for example at a party, particularly the 17-year olds who are less capable of controlling themselves, will more likely still drink alcohol versus the 17-year olds with higher levels of self-control’, highlights Defoe.

But also one’s culture could have an effect on what type of risk exposures he/she will be faced with. ‘For instance, smoking cigarettes is somewhat more of a taboo on Sint Maarten compared to the Netherlands. And thus it is conceivable that on Sint Maarten, youth will encounter fewer chances to smoke cigarettes than youth in the Netherlands, and this is expected to have an impact on the levels of youth smoking. However, culture is overlooked in criminology models on risk exposure and in psychology models on youth risk-taking development.’

’Surprisingly, there appears to be no psychological theory that jointly takes these age-dependent and culture-dependent aspects of risk exposure into account, and/or considers how an ecological factor such as risk exposure might interact with individual factors such as self-control to predict youth risk behavior development’, concludes Defoe. ‘By including these factors in the DNERM framework, the model can also support interventions targeting risk behaviors among youth.’


Ivy N. Defoe, 2021, Towards a Hybrid Criminological and Psychological Model of Risk Behavior: The Developmental Neuro-Ecological Risk-taking Model (DNERM), in Developmental Review.

Dr. I.N. (Ivy) Defoe

Faculty of Social and Behavioural Sciences

Programme group: Forensic Child and Youth Care