The actomyosin cytoskeleton generates mechanical forces that power a range of crucial cellular processes, such as cell migration, cell division, and mechanosensing. Actomyosin self-assembles into contractile bundles that underlie force generation in cells, such as stress fibers. The central element in the assembly of contractile bundles is the molecular motor non-muscle myosin II (NMII), which forms myosin filaments to slide actin filament bundles to generate contractile stress. While it is well-understood how myosin filament assembly is regulated, myosin filaments rarely exist as a single entity in cells. Instead, multiple myosin filaments self-assemble into clusters with their head domains in close contact with each other. While recent studies have characterized the dynamics of myosin clusters at the leading edge of the cell, how cells control the growth of myosin clusters on stress fibers has yet to be carefully characterized. In this thesis work, I studied the dynamics of myosin clusters to understand how they are regulated on stress fibers. First, we visualized and quantitatively analyzed the growth of myosin clusters on stress fibers. Myosin cluster growth is driven by both Rho-kinase (ROCK) activity and myosin motor activity, either through the net association of myosin monomers or filaments to existing clusters or through the coalescence of neighboring clusters. A toy model of myosin cluster growth with myosin self-affinity allows us to recapitulate experimentally-determined myosin cluster sizes. Critically, myosin cluster sizes are determined by the limiting pool of available myosin and the underlying F-actin architecture. Lastly, we extracted the material response functions of stress fibers to myosin-induced strain directly from myosin cluster dynamics, revealing different contractile modalities in different subcellular actomyosin networks. Collectively, our findings provide new insights into the cellular regulation of contractile actomyosin structures.