Proceedings of the Annual Meeting of the Cognitive Science Society

The Linguistic Niche Hypothesis suggests languages adapt complexity to learner populations. However, languages also complexify under certain conditions, and the mechanisms behind re-complexification remain computationally underspecified within this framework. We propose a pressure-driven elaboration mechanism where child learners actively extend morphological patterns, with elaboration strength scaling inversely to current complexity levels. To account for demographic transitions, the model varies the proportion of child and adult learners within each cohort. Using an agent-based transmission model, we compared this mechanism against constant and no-elaboration alternatives. Results indicate that while constant elaboration leads to high complexity, pressure-driven elaboration stabilizes at intermediate levels, a pattern more consistent with documented post-creole languages. The model further predicts asymmetric dynamics, where simplification occurs significantly faster than recovery, and incomplete recovery to original substrate complexity. These findings provide a formal, quantitative framework for understanding how demographic transitions and functional pressures interact to shape linguistic structure across many generations.