RESUMEN
This paper describes a safe method for long term intracerebroventricular (i.c.v.) drug administration. Employing a non-preferred lever to control the self-administration the technique offers advantages over existing experimental methods. To check for any innate preference for one of two levers, male Wistar rats were allowed during the training procedure, to press two levers (L1 and L2) for one hour/day to obtain water as reinforcer for one week in a continuous reinforcement schedule (CRF). One week after surgery, during which a double-guide stainless steel cannula was inserted into both lateral ventricles, rats received 2 microliter of sterile cerebrospinal fluid (CEPH) each time they pressed one of two levers during the daily one-hour session. When a stable baseline was reached, rats were divided into three groups on the basis of their lever preference, and submitted to the testing procedure. A potent mu-opiate receptor agonist, etonitazene (0.1-0.2-1 microgram/infusion), was always associated with the non-preferred lever for each rat. When no obvious preference was shown for either lever the opiate was firstly delivered by L1(for 11 days) and then by L2 (for 20 days). The results indicate that, regardless of which lever had been preferred initially, the rats increased the pressing of only the lever associated with the opiate. The daily amount taken increased linearly and was behaviorally active. This model highlights for the first time the reinforcing properties of drugs given i.c.v. in a free-choice situation.