The nature of the sensory threshold and approaches to solving the threshold problem in psychophys-ics are discussed. Experimental results are presented, as well as empirical facts that do not fully correspond to existing psychophysical theories and do not receive an explanation in line with modern approaches. It was con-cluded that the explanation of the thresholds without description of the participation of consciousness raises doubts. A new look at the nature of thresholds is proposed, in which the threshold is considered as an inevi-table consequence of the processes of classification and categorization. Due to the fact that the operations of identification of the non-identical and discernment of the indiscernible are mandatory in the act of cognition, a zone is formed, within which objectively different elements are identified in consciousness and relate to the same class. This zone exists across the entire spectrum of cognitive tasks, including detection and discrimina-tion. Thresholds fix the boundaries of this zone. Thus, the problem of the threshold requires a description of the work of not so physiological as cognitive mechanisms for its solution.