Mirtazapine and ketanserin alter preference for gambling-like schedules of reinforcement in rats

Abstract

Drug and behavioral addictions have overlapping features, e.g., both manifest preference for larger, albeit costlier, reinforcement options in cost/benefit decision-making tasks. Our prior work revealed that the mixed-function serotonergic compound, mirtazapine, attenuates behaviors by rats motivated by abused drugs. To extend this work to behavioral addictions, here we determined if mirtazapine and/or ketanserin, another mixed-function serotonin-acting compound, can alter decision-making in rats that is independent of drug (or food)-motivated reward. Accordingly, we developed a novel variable-ratio task in rats wherein intracranial self-stimulation was used as the positive reinforcer. Using lever pressing for various levels of brain stimulation, the operant task provided choices between a small brain stimulation current delivered on a fixed-ratio schedule (i.e., a predictable reward) and a large brain stimulation delivered following an unpredictable number of responses (i.e., a variable-ratio schedule). This task allowed for demonstration of individualized preference and detection of shifts in motivational influences during a pharmacological treatment. Once baseline preference was established, we determined that pretreatment with mirtazapine or ketanserin significantly decreased preference for the large reinforcer presented after gambling-like schedules of reinforcement. When the rats were tested the next day without drug, preference for the unpredictable large reinforcer option was restored. These data demonstrate that mirtazapine and ketanserin can reduce preference for larger, costlier reinforcement options, and illustrate the potential for these drugs to alter behavior.

Keywords: Cost/benefit decision-making; Gambling; Intracranial self-stimulation; Serotonin; Variable-ratio.

Figure 1

Flow chart showing contingencies for the free-choice trials in the variable-ratio task. Definitions: VR, variable ratio; FR, fixed-ratio. Free-choice trials allow the rat to make a selection between the VR lever, which delivers the large reinforcer (100Hz) immediately after the required number of presses is completed and the FR lever that delivers the small reinforcer (50Hz) immediately after 3 presses. If no selection is made within 28s, the levers are withdrawn, and a new trial begins 2s later.

Figure 2

Rats press for the positive reinforcer, independent of lever presentation. (A) Lever-pressing in Phase 2. Presses during training for FR3 when only one lever is extended (either left or right). Rats (n=15) self-stimulated equally, independent of which lever was extended. (B) Lever-pressing during Phase 3. Both levers were extended, and the rats selected the preferred reinforcer. Data shown verify that rats (n=14) preferred the large reinforcer when associated with either the right or left lever. (C) Rats (n=11) exerted high physical effort to obtain the large reinforcer in a forced VR10. A similar number of reinforcers were delivered, regardless of the lever assignment. On the left y-axis is the number of lever presses averaged across two consecutive sessions. On the right y-axis is the average number of large reinforcers received. Paired t-test, p>0.05.

Figure 3

Pretreatment with mirtazapine decreased preference for the gambling-like lever by ~20%, switching preference to the FR lever. This switch suggests that the drug was able to attenuate preference for the lever associated with the large reinforcer. In the absence of mirtazapine (i.e., 24h post-injection), preference for the VR lever was restored. The range of the VR schedule was VR6–VR18. One-way repeated measures ANOVA with post hoc Newman-Keuls. **, p<0.01.

Figure 4

Pretreatment with vehicle (A), ketanserin at 1.0mg/kg (B), or ketanserin at 2.5mg/kg (C) had no effect on preference for the VR lever. Ketanserin at 5.0mg/kg (D) decreased preference for the VR lever by ~25% while also increasing the number of omitted trials. In the absence of ketanserin (i.e., 24hrs post-injection), preference for the VR lever was restored. The range of the VR schedule was VR8–VR18. One-way repeated measures ANOVA with post hoc Newman-Keuls. **, p<0.01.