This project is aimed at the investigation of horizontal links in two clusters of constructions featuring the auxiliarized Dutch verbs komen ‘to come’ and krijgen ‘to get, receive’. Horizontal links, i.e. links between constructions which are not in a mother-daughter relation, constitute a topical issue in present-day construction grammar, but their nature and effects, as well as their interplay with other kinds of links that potentially structure the constructional network, are still far from understood. Through a combination of extensive synchronic and diachronic corpus investigation with a series of priming experiments, the project addresses four related clusters of research questions, viz. (i) What is the cognitive plausibility and nature of horizontal links?; (ii) What is the diachronic impact of horizontal links, i.e. in which ways can constructions connected through a horizontal link become more or less similar over time?; (iii) What is the synchronic impact of horizontal links on language processing and grammatical usage?; and (iv) What is the relation/interplay between horizontal links between schematic constructions on the one hand and less schematic, lower-level constructions and lexical items on the other hand? In investigating these clusters of issues, we will systematically distinguish between links rooted in semantic similarity and links rooted in formal similarity.