The physical origin of enhanced star formation activity in interacting galaxies remains an open question. Knowing whether starbursts are triggered by an increase in the quantity of dense gas or an increase in the star formation efficiency therein would improve our understanding of galaxy evolution and make it possible to transfer the results obtained in the local Universe to high-redshift galaxies. In this paper, we analyze a parsec-resolution simulation of a model of interacting galaxies similar to the Antennae Galaxies. We find that the interplay of physical processes such as tides, shear, and turbulence shows complex and important variations in time and space, but that different combinations of these processes can produce similar signatures in observable quantities such as the depletion time and CO emission. Some clouds within the interacting galaxies exhibit an excess of dense gas (> 10⁴ cm⁻³), while others only attain similarly high densities in the tail of their density distribution. The clouds with an excess of dense gas are found across all regions of the galaxies, but their number density varies between regions due to different cloud assembly mechanisms. This translates into variations in the scale dependence of quantities related to cloud properties and star formation. The super-linearity of the relationship between the star formation rate and gas density implies that the dense gas excess corresponds to a decrease in the depletion time, and thus leads to a deviation from the classical star formation regime that is visible up to galactic scales. We find that the α_CO conversion factor between the CO luminosity and molecular gas mass exhibits stronger spatial than temporal variations in a system like the Antennae. Our results raise several caveats for the interpretation of observations of unresolved star-forming regions, but also predict that the diversity of environments for star formation will be better captured by the future generations of instruments.