Photoreactive self-healing semiconductors with suitable bandgaps for solar energy conversion offer an intriguing path to making resilient and lowcost photovoltaic devices through the introduction of a self-recovery path. However, only a few inorganic photovoltaic materials have such quality, and the underlying chemical properties that enable it are unknown, which poses a significant limit to our ability to study and discover new self-healing semiconductors. Recently, we have found that antimony trichalcogenide (Sb 2 Se 3 and Sb 2 S 3 ) and chalcohalides (e.g., SbSeI) can undergo a reversible photoinduced phase transition (PIPT) in which the structure is restored after photoinduced damage incurs to the materials. This group of materials offer a unique opportunity for studying PIPT and its limits. In particular, this group of materials facilitate the study of functional permutation to specific crystalline sites and to finding the limits of PIPT occurrence, which sheds light on the origin of the PIPT and self-recovery of this class of materials. Using Raman spectroscopy of thin films, and following signature vibrations of transition species, we have found that the PIPT magnitude decays upon gradual Bi Sb(1) substitution in a Sb 2−x Bi x Se 3 homologous series, until nearly one in five Sb ions is substituted with Bi. Then, the PIPT diminishes completely. The homologous series occurs along a transition from covalent to metavalent chemical bonding. By expanding our search, we find that a correlation between bonding type and photoreactivity does exist but conclude that it is an insufficient condition. Instead, we suggest, based on bond order and additional DFT calculations, that sufficient bonding states at the bottom of the conduction band are also required. This joint experimental and computational study pushes the limits of designing self-healing inorganic semiconductors for various applications and provides tools for further expansion.