Abstract Detailed aspects of the transition from the solitarious to the gregarious phase in the framework of locust ecology are undoubtedly most important for understanding locust phase polyphenism. Nevertheless, due to obvious difficulties in studying the solitarious phase in nature, such information is limited and mostly available from research carried out under laboratory conditions. In the current study, we examined the dispersal patterns of newly hatched locust nymphs in a laboratory setup that simulated seminatural conditions. This was carried out with no previous manipulation of the nymphs other than controlling their parental density. We comparatively tested the spatial distribution of newly hatched nymphs on perches located at different ranges within an emergence arena, and the expected Poisson (random) distribution. Hatchlings were found to disperse among the perches in a pattern significantly different from that expected by random. Irrespective of their parents’ phase, the observed distributions of all nymphs were clearly clumped, similar or close to those expected for gregarious locusts. It seems that rather than emerging with a parentally derived and predetermined phase, hatchlings have an independent default or innate behavioral state, which reflects at least tolerance if not attraction to conspecifics. The typical phase behavior may later become dominant under the appropriate environmental conditions. These results imply novel perspectives on locust phase transformation, which contribute to our understanding of the formation of locust crowds under field conditions. These should be considered in any rationale for developing a preventative management strategy of locust populations.