Dopamine signaling is thought to play a critical role in reward-related learning and the predominant hypothesis proposes that dopamine neurons signal the expected reward values of sensory stimuli in the form of a short-latency phasic burst. Thus, the reward prediction hypothesis assumes that information relevant to determining stimulus identity informs dopamine neuron firing. Recently, this critical assumption has been questioned and given rise to an alternative hypothetical framework proposing that dopamine activity represents a more generic signal of stimulus salience. Proponents of this competing hypothesis note that the requisite feature-based discrimination would not be possible in a time frame short enough to inform the dopamine response. Instead, the """"""""saliency hypothesis"""""""" proposes that short-latency dopamine activity reflects the relative salience of stimuli (independent of their reward value) and that this signal likely derives from the superior colliculus (SC), a subcortical structure lacking the capacity for fine feature discrimination, but specialized for detecting and locating salient sensory events. Support for this latter view comes from recent anatomical evidence for a projection from the SC to substantia nigra pars compacta and physiological findings showing that SC activity can drive phasic dopamine activity. The presently proposed study provides a critical test of these competing hypotheses by recording from dopamine neurons in the context of behavioral tasks capable of dissociating activity relating to stimulus salience from that which predicts reward value. Understanding the mechanisms of dopamine's involvement in reward related learning, as this proposal aims to accomplish, will assist the explanation of clinical outcomes of diseases related to altered levels of brain dopamine and offer insights for more effective treatment. Several diseases are associated with altered functional levels of brain dopamine. Schizophrenia, Parkinson's disease (PD), attention deficient hyperactivity disorder (ADHD), and the addictions are some examples. The variety and prevalence of dopamine-related diseases makes the knowledge of how dopamine contributes to reinforcement learning particularly timely.