We found striking dissociations between gamma-50 and gamma-80 power in (1) their distribution relative to reward locations on the track, and (2) their evolution with experience. Gamma-50 increased sharply following reward receipt, while gamma-80 power showed a ramp-like increase up to reward receipt, followed by an abrupt decline. Overall gamma-80 power declined with experience, whereas gamma-50 power increased. Kalenscher et al. (2008) reported reward-related oscillations in rats running a lap-based task, but they did not distinguish between gamma-50 and gamma-80. Tort et al. (2008) reported the presence of gamma-50 and gamma-80 in dorsal striatum; while gamma-50 on their T-maze task was highest at the reward site, consistent with the present results, their gamma-80 did not appear to show anticipatory ramping activity. To what extent this reflects task differences or dorsal-ventral differences is not known. It is important to note that on our task, neither gamma-50 or gamma-80 modulation can be straightforwardly explained by a relationship to running speed. For gamma-80, the reward-approach ramp observed on the segment between the final choice point and the first reward site (T4-F1) is accompanied
