Serotonin and tolerance to delay of reward in rats

Tolerance to delay of gratification, taken to reflect impulsiveness, has been proposed to be under the preferential control of central serotonin (5-HT) processes.

The present study further examined the effects of drugs which directly or indirectly alter 5-HT transmission, on behaviour controlled by a delayed positive reinforcer.

Rats were given the choice in a T-maze between two magnitudes of reward: small (two food pellets) and immediate versus large (ten pellets) but delayed. When a 15-s waiting period was imposed in the arm leading to the large reward, rats selected this arm on 65-70% of the trials. This frequency was reduced to less than 40% when the large reward was delayed by 25 s.

In rats whose ascending 5-HT pathways had been lesioned by infusion of 5,7-dihydroxytryptamine (5,7-DHT) into the dorsal raphe, the introduction of the 15-s delay contingency resulted in a transient larger reduction of the frequency of choice of the now-delayed reward, compared to sham operated controls. In contrast, choice behaviour of rats given 5,7-DHT into the substantia nigra did not differ from controls. para-Chlorophenylalanine (pCPA, 150 mg/kg IP, daily for 3 days), a 5-HT synthesis inhibitor, bretazenil (0.5-8 mg/kg IP), a benzodiazepine (BZD) receptor partial agonist, and muscimol (0.25-1 mg/kg IP), a GABA(A) receptor agonist, induced a shift toward immediate reward. In contrast to the other BZDs, alprazolam (1-2 mg/kg IP) enhanced the frequency of choice of the large-but-25 s-delayed reward. Similar increased preference for the large-but-delayed reward was induced by the selective 5-HT reuptake inhibitors, fluoxetine (4-8 mg/kg IP) and fluvoxamine (4 mg/kg IP). The full 5-HT(1A) receptor agonist, 8-OH-DPAT (0.015-0.5 mg/kg IP) enhanced the frequency of choice of the large reward delayed by 25 s, whereas the partial agonists, buspirone (1-4 mg/kg IP), ipsapirone (0.5-1 mg/kg IP) and MDL 73005EF (1-2 mg/kg SC), and the antagonist, WAY 100635 (4 mg/kg SC), reduced the number of choices of the large reward delayed by 15 s. Unexpectedly, WAY 100635 (2 mg/kg), which had no effect on choice whatever the delay, did not counteract the increased tolerance to delay induced by 8-OH-DPAT (0.06 mg/kg) and further reduced the frequency of choice of the large-but- 15 s-delayed reward induced by ipsapirone (0.5 mg/kg).

These effects on tolerance to delay may be accounted for by a subtle balance between the opposing functional consequences of pre- versus post-synaptic 5-HT(1A) receptor activation or blockade. Overall, the present results provide further support to the idea that 5-HT processes participate in the control of impulsive-related behaviour, as assessed from tolerance to delay of reward in this particular T-maze procedure.