The basal ganglia are the key circuitry controlling movement and reward behavior. Both locomotion and reward-related behavior are also modified by dopaminergic input from the substantia nigra and the ventral tegmental area (VTA). If the basal ganglia are severed by lesion or in disease, such as in Parkinson’s disease, the affected individuals suffer from severe motor impairments and often of affective and reward-related symptoms. The subthalamic nucleus (STN) is a glutamatergic key area of the basal ganglia and a common target for deep brain stimulation in Parkinson’s disease to alleviate motor symptoms. The STN serves not only motoric, but also limbic and cognitive functions, which is often attributed to a tripartite anatomical subdivision. However, the functional output of both VTA and STN may rely more on intermingled subpopulations than on a strictly anatomical subdivision. In this doctoral thesis, the role of subpopulations within and associated with the basal ganglia is addressed from both a genetic and a behavioral angle. The identification of a genetically defined subpopulation within the STN, co-expressing Paired-like homeodomain transcription factor 2 (Pitx2) and Vesicular glutamate transport 2 (Vglut2), made it possible to conditionally reduce glutamatergic transmission from this subgroup of neurons and to investigate its influence on locomotion and motivational behavior, giving interesting insights into the mechanisms possibly underlying deep brain stimulation therapy and its side-effects. We address the strong influence of the Pitx2-Vglut2 subpopulation on movement, as well as the more subtle changes in reward-related behavior and the impact of the alterations on the reward-related dopaminergic circuitry. We also further elucidate the genetic composition of the STN by finding new markers for putative STN subpopulations, thereby opening up new possibilities to target those cells genetically and optogenetically. This will help in future to examine both STN development, function in the adult central nervous system and defects caused by specific deletion. Eventually identifying and characterizing subpopulations of the STN can contribute to the optimization of deep brain stimulation and help to reduce its side-effects, or even open up possibilities for genetic or optogenetic therapy approaches.