Are adolescents more optimal decision‐makers in novel environments? Examining the benefits of heightened exploration in a patch foraging paradigm

Abstract

Adolescence is a period of heightened exploration relative to adulthood and childhood. This predisposition has been linked with negative behaviours related to risktaking, including dangerous driving, substance misuse and risky sexual practices.However, recent models have argued that adolescents' heightened exploration serves a functional purpose within the lifespan, allowing adolescents to develop experiential knowledge of their surroundings. Yet, there is limited evidence that heightened exploration in adolescence is associated with positive outcomes. To address this, the present pre-registered study utilised a foraging paradigm with a sample of adolescents aged 16-17 (N = 68) and of adults aged 21 and above (N = 69). Participants completed a patch foraging task, which required them to choose between exploiting a known resource which gradually yields fewer rewards, and exploring a novel, unknown resource with a fresh distribution of rewards. Findings demonstrated that adolescents explored more than adults, which -in the context of the current task-represented more optimal patch foraging behaviour. These findings indicate that adolescents' heightened exploration can be beneficial, as they were able to effectively navigate unknown environments and accrue rewards more successfully than adults. This provides evidence that heightened exploration in adolescence, relative to adulthood, can lead to positive outcomes and contributes to our understanding of the role increased novelty-seeking plays at this point in the lifespan.

Delay discounting and under-valuing of recent information predict poorer adherence to social distancing measures during the COVID-19 pandemic

“…Consistent with previous research, participants adjusted the point that they left patches according to the richness of the environment 32 , 42 . Specifically, participants explored more in the richer environment and explored less in the poorer environment.…”

“…Participants completed two unique environments or ‘orchards’, which differed in the number of rewards available. The availability of rewards was manipulated through changing the rate at which apples depleted from patches, which was based on previous research 42 . Each environment was presented for five minutes with the order of presentation counterbalanced across participants.…”

Reinforcement Learning and Bayesian Inference Provide Complementary Models for the Unique Advantage of Adolescents in Stochastic Reversal

“…This research fills a knowledge gap regarding the adolescent development of reversal learning (also see Hauser et al, 2015;Javadi et al, 2014;van der Schaaf et al, 2011), inspired by rapidly-expanding research highlighting the developmentally-unique role of adolescence across socio-emotional and cognitive contexts (Dahl et al, 2018;DePasque and Galván, 2017;Lloyd et al, 2020;Lourenco and Casey, 2013;Sercombe, 2014), and by the non-linear development of neural and endocrine systems underlying reversal learning (Blakemore et al, 2010;Braams et al, 2015;Giedd et al, 1999;Piekarski, Johnson, et al, 2017;Somerville and Casey, 2010;Sowell et al, 2003;Toga et al, 2006).…”

“…Furthermore, adolescents might explore differently (Gopnik et al, 2017;Lloyd et al, 2020;Somerville et al, 2017) or might be more persistent, a behavioral pattern commonly linked to the PFC (Kehagia et al, 2010;Morris et al, 2016), which continues maturation during adolescence (DePasque and Galván, 2017; Giedd et al, 1999;Toga et al, 2006). Whereas the previous hypotheses targeted the "updating" step of decision making, these two concern the "choice" step, and can be tested in both RL and BI frameworks.…”

“…In humans, adolescence often involves an expansion of environmental contexts and increasingly frequent transitions between them (contextual volatility; e.g., new pastime activities, growing relevance of peer relationships; Albert et al, 2013;Somerville et al, 2017), as well as increased exposure to uncertainty (outcome stochasticity; e.g., increased risk-taking and sensation seeking, increased unpredictability of social interactions; Romer and Hennessy, 2007;van den Bos and Hertwig, 2017). Accordingly, it has been argued that adolescent brains and minds are specifically adapted to contextual 2 volatility and outcome stochasticity, showing an increased ability to learn from and succeed in these situations (Dahl et al, 2018;Davidow et al, 2016;Johnson and Wilbrecht, 2011;Lloyd et al, 2020;Lourenco and Casey, 2013;Sercombe, 2014).…”