Is grammatical complexity shaped by sociodemographic and sociolinguistic factors? The previously accepted "linguistic niche hypothesis" claims that with an increased number of nonnative speakers in a social group (high exotericity), grammatic complexity decreases; on the other hand, grammatical complexity increases amongst isolated linguistic communities (low exotericity). Through the use of spatiophylogenetic modelling of 1314 languages, the authors of this study do not find adequate evidence to support the linguistic niche hypothesis. Instead, they suggest that linguistic complexity is better predicted by phylogeny and geographic contiguity.

Is grammatical complexity shaped by sociodemographic and sociolinguistic factors? The previously accepted "linguistic niche hypothesis" claims that with an increased number of nonnative speakers in a social group (high exotericity), grammatic complexity decreases; on the other hand, grammatical complexity increases amongst isolated linguistic communities (low exotericity). Through the use of spatiophylogenetic modelling of 1314 languages, the authors of this study do not find adequate evidence to support the linguistic niche hypothesis. Instead, they suggest that linguistic complexity is better predicted by phylogeny and geographic contiguity.

The "equi-complexity hypothesis" suggests that there is an equal complexity across languages, meaning that there are constant trade-offs between different domains. Using phylogenetic modelling in a sample of 244 languages, this study follows a diachronic perspective to explore if there is an inversed coevolution within the grammatical coding of nominal and verbal domains. The results show that while there appears to be a coevolutionary relationship between some features of these two domains, there is no evidence to support the idea that all languages maintain an overall equilibrium of grammatical complexity. Rather, the correlation nominal and verbal domains vary between lineages. Austronesian languages do not show a coevolution between the domains. Sino-Tibetan languages seem to have a positive correlation while Indo-European languages appear to have a negative correlation, meaning that this inverse coevolution can be lineage specific.