High-speed chronoamperometry and monoamine-selective electrochemical probes were used to monitor, during each of 5–6 consecutive daily sessions, changes in dopamine (DA)-related oxidation current in nucleus accumbens (NAcc) of rats lever-pressing for a food reward. In trained animals, the first lever-press of each session was preceded by a gradual increase in the electrochemical signal and was followed, during the period the animals retrieved and consumed the food pellet, by a further increase that peaked momentarily when the animal pressed the lever again. This pattern of increases accompanied the initial 1–3 lever-presses of each session after which biphasic changes in electrochemical signal began to emerge. In these cases, each lever-press was preceded by a gradual increase in signal that peaked at the moment the animals pressed the lever and was followed by an abrupt decrease as the animals retrieved and consumed the food pellet. The signal would then start to increase gradually again to peak at a slighly higher level at the moment of the next lever-press. Thus, during the initial part of the session there was a net increment in signal with each lever-press which resulted ina gradual overall elevation of the signal. The increments, however, became progressively smaller as the decrease in signal that followed each lever-press became more pronounced; this slowed the overall rate of increase of the signal until it eventually reached a plateau and remained relatively stable at that level as long as the animals ate earned food pellets. On several occasions, lever-presses were reinforced by twice the usual amount of food. In these cases, lever-presses were preceded by similar increases but were followed by more pronounced and longer-lasting decreases in electrochemical signal. In contrast, non-reinforced lever-presses were followed by less pronounced decreases in signal which then increased more rapidly than was observed after reinforced lever-presses. Toward the end of the session, animals would often ignore earned pellets of food. In these cases, no increases preceded and gradual, rather than abrupt, decreases in signal followed each lever-press. Eventually, the animals ceased to lever-press entirely and this period was accompanied by a gradual return of the electrochemical signal toward baseline values. The tonic elevation of DA levels suggested by the present electrochemical results is in general agreement with previous reports of increased DA efflux in NAcc of animals engaged in feeding and feeding-related behaviors. However, the present results suggest that there are additional phasic changes in DA efflux that are time-locked to each operant response. It would appear that increased DA transmission in NAcc occurs in response to stimuli closely associated with food presentation or in conjunction with behaviors directed toward food, while a transient cessation of DA transmission is associated with the period generally thought to be critical for positive reinforcement of learned behaviors.