A central function of the brain is the ability to predict the timing of future events of behavioral importance based on past experience.  By relating predictive neural activity to behavioral outcome, brain reinforcement systems are thought to mediate changes in synaptic efficacy underlying this ability.  Exemplifying this process is “reward timing activity“ in V1 of rodents, wherein pairing visual stimuli with delayed reward leads to stimulus-specific activity predicting the time of expected reward.  Having modeled how a network learns to produce reward timing activity, we tested and demonstrated that cholinergic innervation of V1 is necessary and sufficient for cued-interval timing to form, and show that such activity is behaviorally-relevant.  Finally, a general theory of how time’s cost is assessed and incorporated into decision making is presented, rationalizing a number of well established, yet curious, observations.