Alternating-treatments design refers to a single-subject design in which two (2) or more treatment conditions are randomly alternated from one observation to the next. Alternating-treatments design is also known as a Discrete-trials design.