The availability of alternative reinforcers can reduce drug self-admin
istration. This 21-day residential study investigated the effect of mo
netary alternatives on marijuana self-administration. Three female and
four male participants performed computer tasks (baseline) in the mor
ning before smoking a sample marijuana cigarette (0.0, 1.8, or 3.9% De
lta(9)-tetrahydrocannabinol (THC)) and receiving the sample alternativ
e ($5.00 voucher). In the afternoon, participants had five opportuniti
es to choose either the marijuana cigarette sampled earlier or a vouch
er for $5.00. Participants were required to meet a criterion level of
task performance to obtain each choice. The monetary performance crite
rion varied from day to day (80, 100, or 120% of baseline); the mariju
ana performance criterion remained constant at 100% of baseline. Choic
es were delivered in the evening, after task completion. Marijuana cho
ice varied as a function of THC concentration and criterion to earn mo
ney. Active marijuana was always chosen more often than placebo, and a
ctive and placebo marijuana mere chosen over money when the criterion
to earn money was high. Task performance improved when criteria were i
mposed, even after participants had smoked the sample marijuana cigare
tte. Subjective ratings of drug effects increased with increasing THC
concentration, but did not predict choice. The availability of a monet
ary alternative was effective in shifting choice to self-administer ma
rijuana, and marijuana choice was sensitive to contingency manipulatio
ns. The results further indicate that contingency manipulations may ov
erride the performance-impairing effects of marijuana observed in othe
r studies.