Economic choice between remifentanil and food in squirrel monkeys

Traditional preclinical research on drug reinforcement relies heavily on self-administration studies conducted in the absence of alternative reinforcers. While effective in assessing abuse potential, this approach lacks predictive power for addiction treatment efficacy. An alternative approach, behavioral economics, focuses on economic choice between drug and non-drug rewards, mirroring real-world scenarios where drug use competes with other valuable activities. Human studies have shown that offering high monetary rewards can reduce cocaine self-administration, illustrating the influence of alternative reinforcers. Contingency management, a therapeutic application of this principle, proves effective in treating substance use disorders. Nonhuman primate models, particularly macaques, offer valuable insights into drug reward and neuroeconomics; however, their cost and complexity limit sample sizes. Squirrel monkeys present a compromise, offering increased statistical power compared to macaques while retaining the advantages of a primate model. This study aims to establish a squirrel monkey model of economic choice between remifentanil (a short-lived opioid) and sweetened condensed milk, using a touchscreen apparatus that varies reward magnitudes parametrically. The goals are to establish a robust behavioral paradigm, demonstrate the economic nature of choice behavior, and evaluate the effects of pharmacological manipulations on reward preference.

The literature supports the limitations of traditional self-administration studies in predicting treatment efficacy. Studies using concurrent choice procedures in humans and nonhuman primates have demonstrated that the availability of alternative rewards significantly impacts drug choice. These studies highlight the importance of considering the economic value of both drug and non-drug rewards in addiction research. While macaques are frequently used in addiction research, the high cost and complexity associated with their use limits sample size. This study explores the feasibility and advantages of using squirrel monkeys as a more statistically powerful and cost-effective alternative.

Ten adult male squirrel monkeys were trained on a touchscreen task to choose between varying amounts of sweetened condensed milk, represented by red circles. After establishing baseline performance, the monkeys underwent surgery for intravenous catheter implantation to enable remifentanil self-administration. A new stimulus (green triangles) represented different doses of remifentanil. Monkeys were then trained to choose between remifentanil and milk stimuli. The final task consisted of 108 trials, including milk-milk and drug-milk choice trials. Reward contrast, a novel metric, was calculated as the difference in the number of symbols for each reward, proportional to their sum. Logistic regression was used to estimate indifference values (IndV), representing the reward contrast at which choice allocation between drug and milk is equal. Experimental manipulations included reward devaluation (milk satiation and saline substitution for remifentanil) and pharmacological pretreatments (morphine and naltrexone). Data were analyzed using paired t-tests and mixed-model ANOVA. Response latencies were also analyzed to assess motivation.

Response latencies were significantly faster for drug-dominant choices, suggesting increased motivation for remifentanil. Baseline choice behavior was stable over time, indicating no sensitization to remifentanil. Milk satiation significantly increased drug choice, and saline substitution significantly decreased drug choice, demonstrating the economic nature of the choices. Naltrexone (0.1 and 0.3 mg/kg) reduced remifentanil choice, but less so than saline substitution. Morphine (1.0 mg/kg) did not affect choice allocation. The IndV served as a sensitive metric for evaluating alterations in preference, showing consistent effects across different manipulations. The reward contrast metric allowed for a detailed analysis of choice behavior, revealing that smaller doses of remifentanil were chosen more frequently than larger doses within a session, indicating that animals are evaluating both offers dynamically. Squirrel monkeys showed a sigmoidal choice allocation curve which shifted from one reward domain to the other as the relative offers of each shifted, similar to dose response curves seen in studies that use varied drug doses but fixed nondrug offer values.

The findings demonstrate the success of establishing a squirrel monkey model for studying economic choice between an opioid and a natural reward. The use of remifentanil, with its rapid clearance, enabled a high-throughput design, allowing for numerous trials and different reward offers within a single session. The reward contrast metric proved a valuable tool for analyzing choice behavior, providing a more nuanced understanding of the decision-making process compared to traditional methods. The sensitivity of the IndV to reward devaluation and pharmacological manipulations validates the economic nature of the choices and establishes the utility of this model for evaluating new addiction therapies. The observed differences in the effects of saline substitution and naltrexone suggest that the latter may also influence the palatability of the milk reward. The lack of effect of morphine on choice allocation confirms the importance of dependence in studying the effects of opioid agonists on choice behavior.

This study successfully established a robust squirrel monkey model of economic choice between remifentanil and a palatable food reward. The model utilizes a novel reward contrast metric and the short-acting remifentanil, enabling a high-throughput approach for studying reward preference and the impact of pharmacological manipulations. The findings validate the economic nature of choice behavior and suggest the model’s potential for evaluating new addiction treatments. Future research could explore the neural circuitry underlying opioid and food reward using this model, and assess the efficacy of contemporary receptor-targeting approaches.

The study used only male squirrel monkeys, limiting the generalizability of the findings to females. The small sample size of the morphine pretreatment condition warrants further investigation. The potential modulating effects of naltrexone on milk reward palatability should be further explored to clarify the mechanism of its effects on choice behavior. The lack of a true devaluation procedure for remifentanil (other than saline substitution) might impact the complete comparison with the devaluation of milk reward.