Directed active motion of motor proteins is a vital process in virtually all eukaryotic cells.Nearly a decade ago, the discovery of directionality switching of mitotic kinesin-5 motors challenged the long-standing paradigm that individual kinesin motors are characterized by an intrinsic directionality. While several kinesin motors have now been shown to exhibit contextdependent directionality that can be altered under diverse experimental conditions, the underlying mechanism remains unknown. Here, we studied clustering-induced directionality switching of the mitotic kinesin-5 Cin8, using a fluorescence-based single-molecule motility assay combined with biophysical theory. Based on the detailed characterization of the motility of single motors and clusters of Cin8, we developed a predictive molecular model, that quantitatively agrees with experimental data. This combined approach allowed us to quantify the response of Cin8 motors to external forces as well as the interactions between Cin8 motors, and thereby develop a detailed understanding of the molecular mechanism underlying directionality switching. The main insight is that directionality switching is caused by a single feature of Cin8: an asymmetric response of active motion to forces that oppose motion, here referred to as drag. This general mechanism explains why bidirectional motor proteins are capable of reversing direction in response to seemingly unrelated experimental factors including clustering, changes in the ionic strength of the buffer, increased motor density and molecular crowding, and in motility assays. Significance Statement:Kinesin-5 motor proteins perform essential functions in chromosome segregation during mitotic cell division. Surprisingly, several kinesin-5 motors have the ability to reverse directionality under different experimental conditions, which contradicts the long-standing paradigm that individual kinesin motors are characterized by an intrinsic directionality. The mechanism underlying this ability to switch directionality has remained elusive. Here, we combine fluorescence-based motility assays and theoretical modeling to analyze cluster-sizedependent motility of the bidirectional kinesin-5 Cin8. Our results show that bidirectional motors can switch directionality because they exhibit an asymmetric response of active motion to drag. This mechanism explains multiple seemingly unrelated experimental factors that have been shown to cause directionality switching of kinesin motors.