Alternative reinforcer response cost impacts cocaine choice in humans

William W. Stoops, Joshua A. Lile, Paul E.A. Glaser, Lon R. Hays, Craig R. Rush

Research output: Contribution to journalArticlepeer-review

30 Scopus citations


Cocaine use disorders are an unrelenting public health concern. Behavioral treatments reduce cocaine use by providing non-drug alternative reinforcers. The purpose of this human laboratory experiment was to determine how response cost for non-drug alternative reinforcers influenced cocaine choice. Seven cocaine-using, non-treatment-seeking subjects completed a crossover, double-blind protocol in which they first sampled doses of intranasal cocaine (5, 10, 20 or 30. mg) and completed a battery of subject-rated and physiological measures. Subjects then made eight discrete choices between the sampled dose and an alternative reinforcer (US$0.25). The response cost to earn a cocaine dose was always a fixed ratio (FR) of 100 responses. The response cost for the alternative reinforcer varied across sessions (FR1, FR10, FR100, FR1000). Dose-related increases were observed for cocaine choice. Subjects made fewer drug choices when the FR requirements for the alternative reinforcers were lower than that for drug relative to when the FR requirements were equal to or higher than that for drug. Intranasal cocaine also produced prototypical stimulant-like subject-rated and physiological effects (e.g., increased ratings of Like Drug; elevated blood pressure). These data demonstrate that making alternative reinforcers easier to earn reduces cocaine self-administration, which has implications for treatment efforts.

Original languageEnglish
Pages (from-to)189-193
Number of pages5
JournalProgress in Neuro-Psychopharmacology and Biological Psychiatry
Issue number1
StatePublished - Jan 10 2012


  • Alternative reinforcer
  • Cocaine
  • Humans
  • Self-administration


Dive into the research topics of 'Alternative reinforcer response cost impacts cocaine choice in humans'. Together they form a unique fingerprint.

Cite this