The `Interface Hypothesis' (IH) was put forward by Sorace and colleagues as an attempt to account for patterns of non-convergence and residual optionality found at very advanced stages of adult second (L2)acquisition. The IH originally proposed that language structures involving an interface between syntax and other cognitive domains are less likely to be acquired completely than structures that do not involve this interface. At the same time, the IH was extended to bilingual first language (L1) acquisition and to the very early stages of L1 attrition, which exhibit optionality in precisely the same structures: this provides a unifying framework for the study of bilingual language development. This paper selectively reviews the research on the IH, addressing some common misinterpretations and outlining the most recent interdisciplinary developments.