Developmental changes in exploration resemble stochastic optimization
Bibliographic Data
| ID | 4655221 |
|---|---|
| Authors | Anna P Giron (University of Tübingen), Simon Ciranka (0000-0002-2067-9781, Max Planck Institute for Human Development), Emily Schulz (0000-0003-3088-0371, Max Planck Institute for Biological Cybernetics), Wouter Van Den Bos (0000-0002-8017-3790, University of Amsterdam), Azzurra Ruggeri (0000-0002-0839-1929, Central European University), Björn Meder (0000-0002-9326-400X, Max Planck Institute for Human Development), Charley M Wu (0000-0002-2215-572X, Max Planck Institute for Human Development, corresponding author) |
| Year | 2023 |
| Volume | 7 |
| Issue | 11 |
| Pages | 1955-1967 |
| Publication date | 2023-08-17 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Nature Human Behaviour (JOURNAL) |
| Journal identifiers | ISSN: 2397-3374 • E-ISSN: 2397-3374 |
| Publisher | Springer Science and Business Media LLC (PUBLISHER) |
| DOI | 10.1038/s41562-023-01662-1 |
| PMID | 37591981 |
| OpenAlex | W4385955542 |
| Language | EN |
| Citations received | 8 |
| References cited | 75 |
Human development is often described as a 'cooling off' process, analogous to stochastic optimization algorithms that implement a gradual reduction in randomness over time. Yet there is ambiguity in how to interpret this analogy, due to a lack of concrete empirical comparisons. Using data from n = 281 participants ages 5 to 55, we show that cooling off does not only apply to the single dimension of randomness. Rather, human development resembles an optimization process of multiple learning parameters, for example, reward generalization, uncertainty-directed exploration and random temperature. Rapid changes in parameters occur during childhood, but these changes plateau and converge to efficient values in adulthood. We show that while the developmental trajectory of human parameters is strikingly similar to several stochastic optimization algorithms, there are important differences in convergence. None of the optimization algorithms tested were able to discover reliably better regions of the strategy space than adult participants on this task
Ambiguity · Generalization · Machine learning · Mathematical optimization · Optimization problem · Randomness · Statistics · Stochastic optimization · Stochastic process · Child and Animal Learning Development · Cognitive Abilities and Testing · Computer Science · Decision-Making and Behavioral Economics · Engineering · Mathematics · Artificial Intelligence
The structure and development of explore-exploit decision making
Exploration, exploitation, and development
Can Childhood be Used as a Model to Understand the Effects of Psychedelics?
Children Understand How Adults’ Achievement Goals Drive Actions
Explore-exploit behavior in humans as a sequential sampling process
Collective Risk Taking in Adolescents and Young Adults
Children use disagreement to infer what happened
Computational psychiatry and the evolving concept of a mental disorder
Optimization by Simulated Annealing
Fuzzy-trace theory
Cortical substrates for exploratory decisions in humans
Plasma Hsp90 levels in patients with systemic sclerosis and relation to lung and skin involvement
Is Adolescence a Sensitive Period for Sociocultural Processing?
Importance of investing in adolescence from a developmental science perspective
Lifespan Psychology
How to Grow a Mind
Two Lines
Resource-rational analysis
A Stochastic Approximation Method
Cognition in harsh and unpredictable environments
Childhood as a solution to explore–exploit tensions
Most people are not Weird
Hidden talents in context
The influences of described and experienced information on adolescent risky decision making
Adolescent risk-taking in the context of exploration and social influence
From tools to theories
Theoretical propositions of life-span developmental psychology
Generalization guides human exploration in vast decision spaces
Asymmetric reinforcement learning facilitates human inference of transitive relations
| Unique citing works | 8 |
|---|---|
| Citations per year | 4 |
| Citation span | 2024 - 2026 (3) |
| Citation velocity | current |
| Highly cited | No |
| Citation types | Neutral: 6 |