Ko and Neuberg propose an integrative framework combining affordance theory and life history theory to explain how people perceive environmental opportunities and threats differently across the life course. The central idea is that human goals shift according to the stage of development (childhood, mating, parenthood), and with them the way environments are perceived also changes. Thus, the same context may be seen as a resource in childhood, an obstacle in youth, or a support in parenthood. The authors apply this framework to three key environmental dimensions in human evolution: the prevalence of genetic relatives, physical violence, and sex and age ratios, showing how each of these variables poses specific affordances that require different management strategies depending on the stage and sex of the individual. The framework highlights three management mechanisms: cognitive–affective recalibration (adjusting attention and sensitivity to relevant signals), strategic behavioural adaptation (modifying decisions and behaviours), and environmental reconstruction (ranging from altering immediate interactions to migrating to new contexts). In so doing, the article challenges the assumption that environmental effects are static and emphasises psychological flexibility throughout the life course. Moreover, it links social, evolutionary and developmental psychology with behavioural ecology, offering precise and testable hypotheses for future research on human–environment interaction.

Table 1 illustrates how core dimensions of ecological affordances have been theorised but remain only partially operationalised. While factors such as kin prevalence, physical violence, and sex–age ratios are well established as shaping cooperation, threat sensitivity, and mating dynamics, their measurement has largely relied on static surveys or self-reports. Artificial intelligence offers a path forward by enabling dynamic, fine-grained assessments – through graph-based social network analysis, computer vision for behavioural cues, and demographic estimation with convolutional neural networks – thus bridging the gap between theoretical models and empirical observation.

Table 1.

AI in behavioural ecology: overcoming traditional measurement limits

Table 2 highlights how advances in computer vision can provide objective measures of ecological affordances across life stages. Convolutional neural networks, such as ResNet or EfficientNet, allow for the automated detection of individual characteristics relevant to sex–age ratios, while pose estimation frameworks like OpenPose or MediaPipe can capture body positions and interpersonal distances, offering insight into signals of cooperation or aggression. At a broader scale, crowd analysis models quantify flow and density patterns, making it possible to infer social pressures that are particularly salient in adolescence. Together, these tools demonstrate how AI can transform affordance research from reliance on self-report to dynamic, real-time observation.

Table 2.

Overview of AI models and their applications to detecting social affordances

Evolutionary simulations with reinforcement agents make it possible to study how cooperation and competition emerge when individuals interact under resource constraints. Through adaptive learning, agents adjust their behaviour in uncertain environments, capturing the cognitive and affective plasticity required to manage shifting threats or opportunities. Population-level models add another layer by testing life-history hypotheses across multiple interacting variables.

Digital twins of the life course provide a continuous representation of human development within changing ecological and social contexts. Such models can integrate demographic transitions, environmental fluctuations, and social pressures, generating scenarios that span from childhood to old age. Virtual environments further allow the exploration of specific situations – such as peer density in adolescence – and their influence on the perception of opportunities and threats.

Generative modelling techniques enable the construction of counterfactual scenarios. By creating alternative ecological or social settings, these approaches help to examine how different conditions could alter the calibration of cognitive and emotional responses across life stages.

Rethinking cognitive recalibration through the integration of AI into the management of affordances across life stages invites a move from static description to dynamic experimentation. By harnessing computational models, it becomes possible to simulate shifting opportunities and threats, generating testable predictions about how individuals adapt cognitively and emotionally over time. This approach holds clear promise for building a computational evolutionary psychology, yet it also demands caution: data biases, the reduction of complex processes, and the risk of over-interpreting simulations must be carefully managed. With rigorous validation, however, AI can serve as a bridge between theory and lived human development, enriching our understanding of how recalibration unfolds across the life course.