In the present study, orally administered nicotine functioned as a discriminative stimulus in 19 out of 20 healthy volunteers with less than 50 lifetime exposures to nicotine. Discrimination between nicotine and placebo was acquired and maintained at a range of doses (1.0–4.0 mg/70 kg) with at least 80 % accuracy. Of the 18 volunteers who completed the study, 9 significantly chose to subsequently self-administer nicotine (nicotine “choosers”) while 9 significantly chose placebo (nicotine “avoiders”). The choosers reported predominately positive subjective effects of nicotine (e.g., significant increases in alert/attentive, good effects, liking), while avoiders tended to report negative effects (e.g., significant increases in dizzy, upset stomach, disliking). Both choosers and avoiders attributed their choice to the qualitative nature of the nicotine effects. Furthermore, in both ratings and written comments summarizing the reasons for their choices, nicotine choosers indicated their choice was based on subjectively positive effects of nicotine administration in contrast to nicotine avoiders who indicated their choices were based on avoiding subjectively negative effects of nicotine administration. The concordance of qualitative subjective reports with choice or avoidance of nicotine, along with the procedure of requiring volunteers to make a minimum of ten repeated choices between nicotine and placebo, suggests that the designation given to volunteers as a chooser or avoider was not based on chance responding. We believe this is the first demonstration that nicotine can function as a reinforcer in humans who are neither current nicotine users nor have a significant past history of nicotine use from any route of administration.

In contrast to prior studies that showed that nicotine did not function as a reinforcer in never-smokers (Perkins et al. 1997; 2001; Hughes et al. 2000), the present study showed that 50 % of never-smokers demonstrated nicotine choice after acquiring a low-dose nicotine versus placebo discrimination over an average of 42 sessions (range, 10 to 85). In contrast, laboratory studies using similar choice procedures comparing d-amphetamine and placebo in non-drug using participants have shown that d-amphetamine is preferred to placebo in the majority of participants after as few as two exposures each to drug and placebo and in absence of explicit discrimination training (de Wit et al. 1986; 1987; Foltin and Fischman 1991). Thus, although the present study does demonstrate nicotine reinforcement in never-smokers, oral nicotine appears to have substantially less efficacy as a reinforcer than oral d-amphetamine when tested under reasonably similar laboratory conditions.

Although the prevalence of nicotine use disorder (i.e., addiction) is high in the general population, nicotine is an atypical drug of abuse in that its initial subjective effects are often not pleasant and, furthermore, positive subjective effects are not a robust predictor of the development of subsequent addiction (de Wit and Phillips 2012; Haertzen et al. 1983). In the present study, nicotine choosers reported choosing nicotine because of positive subjective effects while nicotine avoiders reported choosing placebo to avoid negative subjective effects. However, it is interesting to note that five of the nine choosers showed a significant increase in at least one negative subjective effect (Table 1, open circles for nicotine choosers), while four of the nine avoiders showed a significant increase in at least one of the positive subjective effects (Table 1, filled squares for nicotine avoiders). These findings underscore the complex role between subjective effects and nicotine reinforcement in never-smokers. Although researchers have described a progression of stages from initial nicotine exposure to nicotine addiction, the empirical evidence for such stage-specific predictors is weak (USDHHS 2010) and the present study does not provide information about the relationship of the initial subjective and reinforcing effects to the possible development of subsequent nicotine use disorder.

Although the sample size was small, there was no compelling evidence that past use of nicotine was a significant determinant of the reinforcing effects of nicotine in this study, as has been suggested by previous studies (Neugenbauer et al. 2014). The volunteers in this study were current nicotine non-users with less than 50 lifetime exposures to nicotine. Of the nine who reported any past exposure, six and three were nicotine choosers and avoiders, respectively.

The study was conducted with oral nicotine administration although inhaled nicotine is the most common route of administration. The present study used the oral route of administration to facilitate double-blind administration procedures and to allow for the administration of exact nicotine doses, which is problematic with inhaled delivery. Oral bioavailability of nicotine is 20 to 45 %, likely because of first-pass metabolism (Hukkanen et al. 2005). Administration of 4 mg of oral nicotine (approximately equivalent to the highest dose administered in the present study) had a C max of 6.4 to 7.5 ng/ml and a T max of 1.3 to 1.5 h (Benowitz et al. 1991; D’Orlando and Fox 2004). In contrast, the bioavailability of inhaled nicotine from cigarette smoke is 80 to 90 %, with a C max of 15 to 30 ng/ml and a T max of 5 to 8 min (Hukkanen et al. 2005).

Several limitations of the study should be noted. Female participants were over-represented. Also, how representative the small population of study participants is of the general population of never-smokers is unknown. Although volunteers were instructed to complete questionnaires in response to timer alarms, real time assessment of subjective effects responses using Ecological Momentary Assessment methods would have been preferable (Stone and Shiffman 2002). A potential concern about the study design is that the procedure of using successive blocks of 10 sessions to establish significant discrimination accuracy or significant choice behavior would have theoretically resulted in significance eventually by chance alone. However, the procedure is very unlikely to account for the present results because significant discrimination and significant choice occurred after a mean of only 12.2 and 11.1 sessions, respectively, in the relevant dose conditions. Therefore, there were too few sessions for significant discrimination or choice to have occurred by chance through this mechanism.

In conducting this study of the potential reinforcing effects of nicotine in volunteers with neither current nicotine use nor significant previous use of nicotine, careful consideration was given to the theoretical risk that, after the study, participants might seek out nicotine and become habitual users. In a discussion of human participant issues in drug abuse research, the College on Problems of Drug Dependence concluded that exposure of drug-naïve individuals to abused drugs in a medically monitored setting is unlikely to create addiction or exacerbate pre-existing risk factors for addiction (College on Problems of Drug Dependence 1995). Furthermore, oral nicotine delivery and other forms non-inhaled nicotine delivery are considered to have a very low abuse potential in never-smokers (Henningfield and Keenan 1993; Houtsmuller et al. 2002; Gerlach et al. 2008, but see Etter 2007). In addition, to further reduce the possibility that volunteers would start using nicotine after the study, volunteers were informed that they could receive a wide range of different substances and they were never debriefed about the study objectives or use of nicotine. To the authors’ knowledge, no participant engaged in tobacco smoking behavior or use of nicotine-containing products during the study or in the 2 weeks immediately following completion of the study.

Future research would benefit from the assessment of plasma and saliva nicotine levels to determine if rate of onset, peak plasma levels, or other metabolic differences are important determinants of nicotine reinforcement in human nicotine non-users as is suggested by studies in animals (e.g., Pastor et al. 2013; Wing and Shoaib 2013). Likewise, an examination of the possible role of genetic polymorphisms, age, gender, and ethnicity as determinants of individual differences in the reinforcing effects of nicotine would be of value (e.g., Morel et al. 2014; Schuck et al. 2014).

Nicotine addiction in the form of cigarette smoking is a leading cause of mortality world-wide (USDHHS 2014; WHO 2013). Improved understanding of vulnerability to nicotine reinforcement in nicotine-naïve individuals may be vital for understanding the development of tobacco addiction, and improving smoking prevention interventions. The rapidly developing technology and expanding marketing of electronic nicotine delivery devices (King et al. 2014), especially to youthful nicotine non-users (McMillen et al. 2014; Vakkalanka et al. 2014) underscores the importance of further research of nicotine reinforcement in nicotine-naïve populations.