Cotton yarn doesn't stretch much. Wool has some give, but nothing like the extensibility of a knitted sweater. The fabric stretches far more than the yarn it's made from. Where does the extra elasticity come from?
Elisabetta Matsumoto and Krishma Singal at Georgia Tech answered this by combining physical experiments with simulations that tracked individual yarn paths through stitches. The elasticity of knitted fabric is an emergent property of stitch topology — the pattern of overcrossings and undercrossings that defines each stitch type. Change the topology and you change the stretch, even with the same yarn.
This is not the same as saying “structure matters.” Everyone knows that a knitted fabric and a woven fabric behave differently. The finding is more specific. Within knitting, the particular topological choices — knit versus purl, the sequence of crossings — program the mechanical response. The topology is not just a pattern. It's a set of mechanical instructions written in yarn.
The result inverts the usual materials-science logic. In most engineering contexts, you choose a material for its properties, then shape it. Knitting works backward: you choose a topology for its properties, then feed any yarn through it. The yarn supplies constraints — friction, bending stiffness, diameter — but the dominant mechanical behavior comes from how those constraints interact at the crossing points. The crossings create a phase space of mechanical responses that the yarn alone doesn't contain.
A stiff yarn knitted loosely can be more elastic than a soft yarn knitted tightly, depending on the stitch pattern. The topology overrides the material. This is emergent in the strict sense: the property exists at the fabric level but not at the yarn level, and it arises from the relational structure between components rather than from the components themselves.