Modelling physical-biological interactions in the early life of fish is becoming an integral part of theoretical and applied marine ecology. A workshop on 'Advancements in modelling physical-biological interactions in fish early-life history' (WKAMF) was held in April 2006 to review recent developments and identify future research directions. Here we provide an overview of the information presented at WKAMF (some of which is published in this Theme Section), discussions that took place at the workshop, and the authors' perspectives as workshop co-Chairs. The major themes identified at the workshop were the need for enhanced model validation and sensitivity methods and improved understanding of physical and biological processes. Using the appropriate level of model complexity required for each model application is important; developing quantitative consistency of model results with good quality observational data is critical. In addition, improving our prediction of physical processes, such as circulation patterns and turbulence, will advance our knowledge of fish early life, as will a better understanding of biological processes like mortality, behaviour, and energetics. The latter stage-dependent, often species-specific, processes pose particular challenges, although technical advances in field and laboratory observations are likely to result in considerable progress in the near future. Finally, there is a clear requirement for interdisciplinary collaboration between modellers, field scientists and laboratory scientists. Studies receiving input from a wide range of disciplines will increase our understanding of fish early-life ecology and prediction of recruitment to fish populations.