The allocation of "choice" behavior provides a measure that may be useful in developing experimental models of clinical relapse. In the present experiments, indirect monoaminergic agonists [cocaine, 1-(2-[bis(4-fluorophenyl)methoxy]ethyl)-4-(3-phenylpropyl)piperazine (GBR 12909), desipramine, and citalopram], and dopaminergic D1 family agonists [(+/-)-6-chloro-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF 82958), R-(+)-6-bromo-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine (R-(+)-6-BrAPB), and 6-chloro-7,8-dihydroxy-3-methyl-1-(3-methylphenyl)-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF 83959)] and D2 family agonists [quinelorane, R-(-)-10,11-dihydroxy-N-n-propylnorapomorphine (R-NPA), (+)-N-propyl-hydroxynaphoxazine [(+)-PHNO], and S-(+)-(4aR,10bR)-3,4,4a,10b-tetrahydro-4-propyl-2H,5H-[1]benzopyrano-[4,3-b]-1,4-oxazin-9-ol (PD 128907)] were evaluated for their capacity to alter the distribution of choice behavior in cocaine-experienced monkeys. Rhesus monkeys responded on two levers (injection-lever and food-lever) under concurrent fixed ratio 30; fixed ratio 30 schedules of i.v. cocaine and food delivery. Under training conditions, the distribution of behavior was related to the unit dose of i.v. cocaine: when saline was available, responding occurred predominantly on the food-lever and when reinforcing doses of cocaine were available, responding occurred predominantly on the injection-lever. Drugs were studied by administering i.m. pretreatment doses before components in sessions of i.v. saline availability. Cocaine produced dose-related increases in injection-lever responding in all monkeys, whereas desipramine failed to alter the distribution of behavior in any monkey. The dopamine transport blocker GBR 12909 and each dopamine D1 family agonist markedly increased injection-lever responding in three of four monkeys; the serotonin transport blocker citalopram and D2 family agonists were comparably effective in only one or two monkeys. These results agree with previous findings of similarity in the behavioral effects of cocaine and indirect or direct dopamine agonists and suggest, furthermore, that i.v. self-administration behavior engendered by priming doses of cocaine may involve actions mediated through both D1 and D2 families of dopamine receptors.