Imaging reward function in healthy volunteers with PET and fMRI

Objectives In animal models, mesolimbic dopamine (DA) release has been associated with reward salience. We tested DA release using [¹¹C]raclopride PET, and neural activity with fMRI, in 12 healthy volunteers undergoing the monetary incentive delay task (MIDT).

Methods Scans were acquired for 40 min, starting 40 min after Bolus / Constant infusion of [¹¹C]raclopride on the HR+camera, at baseline and after MIDT. BP_ND was derived by equilibrium analysis in the ventral striatum (VST), the cerebellum was used as reference, and ▵BP_ND across conditions was measured. Subjects also underwent fMRI with MIDT (2 x12 min) and BOLD signal changes were measured in both sessions.

Results MIDT caused an average ▵BP_ND of 3.9% in VST (p=0.28). The fMRI pattern of activation was highest in the VST. Correlation of VST ▵BP_ND with fMRI BOLD % change during the 1st session between the highest reward condition (win$5) and neutral conditions (win/lose $0), or between win$5 and average of all conditions, yielded a trend level (r=-0.433, p=0.16) or a significant (r=-0.697, p=0.01) finding, respectively.

Conclusions We did not find a significant effect of MIDT on DA release in VST. This may be related to the combination of rewarding and non-rewarding stimuli within the task, or to our delayed scanning session not detecting transient changes in DA release. However, the significant correlation in VST between ▵BP_ND and fMRI BOLD signal is consistent with the notion that reward related activity in the limbic striatum is related to DA transmission.