Association for Behavior Analysis International

Tobacco taxation has been demonstrated to reduce the incidence and prevalence of smoking, but there is inadequate clarity about the relationship between cigarette consumption and the price of cigarettes. Behavioral economic demand curve analysis permits a precise characterization of this relationship and may be useful for optimizing tobacco tax policy. Toward this end, the current study assessed cigarette demand and other relevant demographic and motivational variables in a large sample (N = 1042) of adult smokers (5+/day) recruited from the community. Cigarette demand was assessed using a Cigarette Purchase Task assessing estimated consumption on a typical day at 73 intervals ($0-$10). Analyses revealed prototypic demand, initial inelasticity followed by a period of elasticity and then full suppression. Paired samples t-tests revealed significant differences at every viable price interval. The largest magnitude decreases were at $3.80-$4, $4.80-$5, $5.80-$6, and $6.80-$7, suggesting a 'left digit' effect. These findings suggest that (a) current tobacco prices exist within the elastic portion of the demand curve and prices increases will disproportionately decrease smoking, and(b) price increases that are yoked to transitions from one whole dollar amount to the next will disproportionately reduce smoking. The role of demographic and smoking-related variables will be discussed.

Behavioral economics provides methods for precisely characterizing the relative value of commodities as reinforcers, including addictive commodities such as cigarettes and alcohol. Two examples are the Cigarette and Alcohol Purchase Tasks (CPT, APT), which are validated measures for assessing the relative value of each drug based on changes in individual consumption based on changes in price. However, there are no published studies on the temporal stability of a CPT and only one study on the APT. To address this, the current study examined the test-retest reliability of a CPT and an APT in a community sample of eleven adult smokers at two time points,1 week apart. Reliability was determined by examining correlations between three relevant indices of demand: Intensity (i.e., consumption at the lowest price), OMax (maximum financial expenditure), and PMax (i.e., price at which expenditure is maximized). For the CPT, correlations for intensity, OMax, and PMax were .98, .94, and .93 respectively. Correlations on the APT for intensity, OMax, and PMax were .83, .87, and .80. All correlations were statistically significant. These data suggest that demand for both cigarettes and alcohol remains highly stable over a1 week time period and provides further psychometric validation for these measures.

In recent years, a number of clinical reports have implicated pramipexole, a common component of dopamine replacement therapy for Parkinson's disease (PD), in the development of impulse control disorders in patients with PD. Experimental evaluations of pramipexole's effects on traditional measures of impulsivity in humans (Hamidovic, Kang, & de Wit, 2008; Voon et al., 2010) and nonhumans (Madden et al., 2010) have produced mixed results. However, even in the case of positive findings like those of Madden et al.'s Experiment 1, in which acute pramipexole increased rats' impulsive choices, the specific behavioral mechanism (e.g., delay sensitivity) by which pramipexole modulates choice remains unknown. The present study used the generalized matching law to describe the effects of acute and repeated pre-session pramipexole on rats' allocation of responding in a concurrent-chains schedule. Of particular interest were changes in the slopes of linear regressions obtained from drug sessions relative to those from non-drug or saline sessions (i.e., changes in delay sensitivity). Both regimens of pramipexole dose-dependently decreased sensitivity to differences in reward delay, as indicated by shallower slopes at larger doses; an effect that was reduced with repeated administration yet is largely inconsistent with increased impulsive choice. Alternative explanations for these findings will be discussed in the context of ongoing work related to pramipexole's effects on amount sensitivity and contingency discrimination.

The primary purpose of this study was to further investigate how acute nicotine deprivation affects cigarette smokers' decision making in delay discounting tasks for hypothetical money rewards. To determine whether changes in discounting were the result of changes in sensitivity to reward amount or delay, choice was investigated in four separate tasks: (a) typical delay discounting, (b) fixed vs. variable delays, (c) fixed vs. variable amounts, and (d) delay discounting with reward amounts and delays matched to tasks two and three. Choice in fifteen adult cigarette smokers was examined following a period of ad lib smoking and 24-hours of nicotine abstinence. Ten adult nonsmokers were included as a comparison group and their discounting rates also were measured across two sessions. In the typical delay-discounting task, smokers' showed steeper rates of delay discounting during the nicotine deprivation condition than the ad lib condition whereas nonsmokers' rates remained unchanged across sessions. Findings from the tasks in which reward amount and delay were held constant suggest that the changes in smokers' discounting rates were more likely the result of changes in sensitivity to reward amount than reward delay.

Delay-discounting procedures are thought to measure impulsive choice, a key component of Attention-Deficit Hyperactivity Disorder (ADHD). d-Amphetamine and methylphenidate are indirect dopamine (DA) agonists that are used to treat ADHD. Analogs of benztropine (BZT) are also indirect DA agonists; however, they typically have effects different from other indirect DA agonists, including diminished indications of abuse liability. The current experiment compared the effects of typical (cocaine, methylphenidate, d-amphetamine) and atypical (BZT analogs) indirect DA agonists on delay discounting in rats using two procedural variations. In one variation, the delay to the large reinforcer was varied in a fixed, ascending order within sessions. In the other, the delay to the large reinforcer was varied in random order within sessions. Under the ascending order procedure, discounting was greater than under the random order procedure. Under both procedures, d-amphetamine decreased whereas the BZT analog, AHN 2-005, increased discounting. Cocaine produced minimal changes in discounting. Methylphenidate produced minimal changes in discounting under the ascending procedure and increases and decreases under the random order procedure. The other BZT analogs produced decreases in discounting under the ascending procedure and either increases (JHW 007) or increases and decreases (AHN 1-055) under the random order procedure. These results suggest possible therapeutic uses of some BZT analogs and demonstrate that the effects of drugs on discounting can depend on procedural specifics.

Delayed reward discounting (DRD), a common behavioral economic index of impulsivity, is hypothesized to be a fundamental behavioral process of drug dependence. Recent advances in neuroeconomics have contributed to our understanding of the neurobiology of DRD in healthy samples and individuals with addictive disorders. However, the application of neuroeconomic paradigms to alcohol use disorders is limited. This study used functional MRI (fMRI) to examine the neural correlates of DRD in a sample of heavy drinkers. Males who drank at least 21 drinks per week participated in an event-related fMRI paradigm during which they responded to choices for smaller and larger amounts of money, available immediately or after a delay. Initial results from the firstfive participants indicate that DRD choices elicited significant BOLD activation in a network of cortical and subcortical regions. Impulsive decisions (preference for smaller immediate reward) were associated with amygdala activation and orbitofrontal cortex deactivation. Restrained decisions (preference for larger delayed reward) were associated with generally greater activation across the decision making network. Restrained choices were further supported by bilateral activation of cognitive control regions in lateral prefrontal cortex. These preliminary results are consistent with prior studies of healthy adults, suggesting common neural processing across samples.