Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Sep 9;124(2):e70054. doi: 10.1002/jeab.70054

Polydrug abuse: Choice between drugs as a function of concurrent nonindependent ratio sizes

Richard A Meisch 1, Thomas H Gomez 2, Scott D Lane 1,✉
PMCID: PMC12418298  PMID: 40923342

Abstract

Polydrug abuse is the persistent self‐administration of more than one reinforcing drug. The present study provided rhesus monkeys concurrent access to two drugs: 8% alcohol and solutions of either cocaine or methadone. The liquids were available under concurrent nonindependent fixed‐ratio (FR) schedules across increasing and then decreasing ratio sizes. These schedules generate high rates of changeover responses and yield a dependent variable of responses per delivery that is not rigidly tied to the ratio‐schedule value. The programmed schedule size was equal for both liquids and increased in the sequence 8, 16, 32, and so on until responding decreased, whereupon the schedule size was decreased in reversed order to the original steps. Eight percent alcohol was strongly preferred at the nonindependent FR 8 FR 8 baseline. As schedule size increased, intake of the 8% alcohol solution decreased and intake of the alternative liquid increased. Consumption of the alternative liquid generally remained elevated over initial values when schedule size decreased. The data can be analyzed in several ways, including consumption as a function of price (behavioral economics) and log of relative response rates as a function of log of relative deliveries (matching), thereby providing an interface between behavioral economics and matching analyses.

Keywords: behavioral economics, concurrent nonindependent ratio, drug choice, drug self‐administration, polydrug abuse, rhesus monkeys, schedules

INTRODUCTION

Polydrug abuse is the parallel use of more than one drug. It is common and serious (Cami & Ferré, 2003; Connor et al., 2014; Kedia et al., 2007). Polydrug abuse is harder to treat and has worse outcomes than abuse of single drugs, and it is accompanied by greater psychiatric and other health problems (Connor et al., 2014). Polydrug abuse is strongly associated with overdoses, both fatal and nonfatal, especially when opioids are combined with other drugs (Cheatle, 2015). Fentanyl, in particular, is a component of many polydrug mixtures (Harper et al., 2023; Janssen & Vuolo, 2024; Potoukian et al., 2023).

Many studies of drug combinations use single response procedures; however, methods employing choice are (a) more sensitive to differences in reinforcing effects, (b) less susceptible to direct disruptive effects of drugs, and (c) provide more valid measures of drug taking than do single response procedures (Banks & Negus, 2012; Lamb et al., 2016; Meisch & Lemaire 1988, 1990; Meisch & Stewart 1995; Meisch et al., 1996; Negus, 2005). For example, in one study with rhesus monkeys, methadone, cocaine, and methadone–cocaine combinations were examined (Wang et al., 2001). When the three solutions were studied sequentially, all maintained similar response rates. However, when the combination was present along with either component drug, the combination was preferred. A problem with choice between a drug combination and a component drug is that most polydrug users choose between separate drugs and do not use a premixed combination. Methods using choice are also similar to many human laboratory studies where participants have choices between drug and another reinforcer (Comer et al., 2013; Haney et al., 2011; Walsh et al., 2016).

Separate drugs can be presented under concurrent fixed‐ratio (FR) FR schedules. In some cases, with monkeys, oral consumption of both liquids occurs (Carroll, 1987). In other cases, with the intravenous (IV) route, monkeys showed seemingly random responding (Huskinson et al., 2015). Rats exhibited three patterns of IV self‐administration of concurrent methamphetamine and fentanyl: preference for only one drug, preference for both drugs, or a preference that alternated across sessions (Seaman, Lordson, et al., 2022). When the choice was between equally effective doses of cocaine and MDPV (a synthetic cathinone), there was no preference (Seaman, Rice, et al., 2022). Rats given concurrent oral access to oxycodone and alcohol take some of each drug; however, lack of vehicle control values make it difficult to assess reinforcing effects (Amico et al., 2022). One factor controlling choice is that concurrent FR FR schedules do not generate high rates of changeover responses between schedules, as progress on one schedule does not change the probability of obtaining a reinforcer on the other schedule.

One purpose of the present study was to analyze choice between two reinforcers of unequal effects, in this case between two drugs with different reinforcing effects. Choice was studied as a function of schedule size. Nonindependent FR FR schedules were used because these procedures promote responding between both alternatives (MacDonall, 1988; Meisch & Gomez, 2013). With these schedules, responses on each operandum simultaneously count toward the completion of the response requirements of both ratio concurrent schedules. Thus, responding on one schedule increases the probability of obtaining a reinforcer on the other schedule (Bellow & Lattal, 2023; MacDonall, 1988; Meisch & Spiga, 1998; Shull & Pliskoff, 1971). This arrangement is like concurrent interval schedules where the passage of time contributes simultaneously toward the completion of both interval schedules. In the current study, nonindependent‐ratio (nFR nFR) schedules were employed to generate intake of two separate drugs when one drug was substantially less preferred. Ratio schedules were used because concurrent interval schedules can result in the development of side preferences due to reinforcement of the first response that follows extended pauses (Iglauer & Woods, 1974). A second purpose was to determine the effects of schedule size, or price, on choice. A third purpose was to study the effects of changes in drug intake as function of earlier choices. The study was accomplished by providing three monkeys with choices between alcohol and cocaine or methadone under nFR schedules. Nonindependent FR values were presented in ascending order (nFR 8, 16, 32, 64, or 128) and then descending order (nFR 128, or 64, 32, 16, and 8), across blocks of at least six sessions. The results have possible implications for a quantitative relation between matching and behavioral economics.

METHODS

Subjects

The subjects were three adult male rhesus monkeys (Macaca mulatta). Two had more than 8 years of experience with oral drug self‐administration (Crash and Raja), and one (Lucas) had more than 6 years of experience. Their behavior had been reinforced under standard (independent) concurrent FR FR schedules and under nonindependent concurrent nFR nFR schedules (Meisch & Gomez, 2013, 2016). Crash and Raja had experience with responding reinforced by alcohol, methadone, and cocaine, and Lucas with alcohol and cocaine.

The monkeys were individually housed in their chambers in a climate‐controlled room (22.8°C) with a 12‐hr light–dark cycle (lights on at 0700 hours). Water was provided ad libitum via attached water bottles for 24 hr per day. The monkeys were maintained at 85%–90% of their optimal weights by feeding a daily, measured quantity of commercial chow (High Protein Monkey Diet 5045; Lab Diet, St. Louis, Missouri, USA), fresh fruit, and vegetables. At the start of the study, their weights were as follows: Crash, 10.1, Raja, 9.75, and Lucas, 8.95 kg.

Each monkey's optimum weight was determined by two veterinarians who are familiar with the monkeys and was based on each monkey's health and body condition (cf. Pugh et al., 1999). The food allotment maintained stable weights for the duration of the study. Food restriction is known to increase drug self‐administration (Carroll & Meisch, 1984) and reinforcing effects of drugs (Carr, 2002; Kliner & Meisch, 1989). Food restriction also protects against obesity (Meisch & Lemaire, 1988), lengthens median life span, and benefits general health (Kemnitz, 2011; Mattison et al., 2017). The monkeys were fed 1 hr after the end of each session. Animal care followed the guidelines of the National Research Council (2011) and were conducted at AAALAC‐l‐accredited facilities.

Apparatus

The monkeys were individually housed 24 hr per day in primate cages, which also served as the experimental chambers. A liquid delivery apparatus panel was attached to the outside of one side wall, and spouts and stimulus lights protruded into the cage through holes cut in that wall. On the back of the apparatus panel was a T‐shaped bar. The horizontal portion of this bar was elevated above the level, and on each limb of the horizontal bar was fastened a stainless‐steel reservoir covered with a lid. Delivery of liquid occurred as a function of the pressure differential between the elevated reservoir and the spout, with liquids contained in each reservoir passing through polyethylene tubing to a solenoid‐operated valve at the rear of one of the two brass spouts. These spouts (1.2 cm outside diameter, 0.2 cm inside diameter) extended 2 cm into the cage, 64 cm above the floor, and 15.5 cm on either side of the midline. The spouts were embedded in Plexiglas disks that covered the 7‐cm‐diameter holes in the cage wall through which they entered. At each spout, two 1.1‐W lights, one located 2.5 cm on either side of the spout and visible through the Plexiglas, were aligned diagonally; these spout lights were capped with green translucent lenses. Another two 1.1‐W spout lights, one located 2.5 cm on either side of the spout, were aligned on the opposite diagonal and were capped with white translucent lenses. Thus, each spout was in the center of a square pattern of four spout lights, two green and two white. The electronic components for the drinkometer circuit were housed in an enclosure at the rear of the spout. A 2.5‐cm‐diameter cluster of green light‐emitting diodes was located 11.5 cm directly above each brass spout. The liquid‐delivery apparatus has been described extensively elsewhere (Gieske, 1978; Henningfield & Meisch, 1976). The programming of experimental events and the recording of behavior were accomplished with a Dell computer, MED‐PC software, and Med Associates Inc. interface equipment.

Each reinforced response operated a spout's solenoid for a set duration, typically of 150 ms, allowing gravity‐fed delivery of approximately 0.65 mL of liquid. The exact time of solenoid operation was calibrated for each solenoid, and the duration was controlled by the MED‐PC software and Med Associates Inc. interface equipment. The volume of each liquid consumed in a session was calculated by measuring the differences in the volume in each reservoir before and after each session. The average volume of liquid per delivery was calculated by dividing each volume consumed by the number of deliveries occurring on the corresponding side. To keep the average volume of liquid per delivery at approximately 0.65 mL, the duration of operation of a spout's solenoid was adjusted as necessary across sessions. In four prior studies, blood alcohol levels were determined and the levels confirmed that monkeys consume the alcohol solutions with this procedure (Henningfield & Meisch, 1978; Macenski & Meisch, 1992; Meisch & Lemaire, 1991; Stewart et al., 1996).

Procedure

The experimental conditions and their sequence are listed in Table 1. The monkeys were concurrently presented with two liquids, either 8% alcohol and a drug or water. The nFR schedule size was first increased and then subsequently decreased across blocks of sessions.

TABLE 1.

EXPERIMENTAL CONDITIONS AND THEIR SEQUENCE.

Crash
Condition nFR 8 nFR 16 nFR 32 nFR 64 nFR 128
8% A vs. 0.4 C 8 16 32 64 128
8% A vs. +1% A 8 16 32 64 128
8% A vs. W 8 16 32 64 128
Lucas
Condition nFR 8 nFR 16 nFR 32 nFR 64 nFR 128
8% A vs. 0.4 C 8 16 32 64 128
8% A vs. 0.4 M 8 16 32 64 128
8% A vs. W 8 16 32 64 128
Raja
Condition nFR 8 nFR 16 nFR 32 nFR 64 nFR 128
8% A vs. 0.4 nC 8 16 32 64 128
8% A vs. 0.4 M 8 16 32 64 128
8% A vs. W 8 16 32 64 128
Open in a new tab

Note: 8% A = 8% alcohol; 0.4 C = 0.4 mg/ml cocaine; 0.4 M = 0.4 mg/mL methadone; W = water vehicle; 1% A = 1% alcohol.

Prior to the session

Some of each 250‐mL solution was drained through the tubing leading from the reservoir to ensure that the appropriate solution was present on the first delivery of the session. The remaining volume was remeasured to determine the volume present at the start of the session. After the session, the volume was measured again and the amount consumed was calculated by subtracting this volume from the volume at the start.

Sessions

Experimental sessions were 3 hr in length (from 1000 to 1300 hours) and were conducted 7 days per week. The cluster of green light‐emitting diodes above each spout, which functioned as a discriminative stimulus, blinked at a rate of 10 Hz when liquid was available from a spout during sessions. Identical discriminative stimuli were used for both spouts to control for differential responding that might otherwise result from the presence of dissimilar exteroceptive visual stimuli. Each mouth contact with a spout completed a drinkometer circuit and illuminated the green‐lensed pair of spout lights for the duration of the response. Deliveries of liquids were contingent on completion of a nonindependent fixed‐ratio schedule. The final response in the schedule requirement initiated the liquid flow of approximately 0.65 mL of the appropriate solution. To further reduce possible differential responding that might be caused by a monkey's preference for a particular spout, the drug and vehicle (water), or pairs of drug doses, were alternated between spouts each session.

Reinforcement schedule

Nonindependent concurrent fixed‐ratio fixed‐ratio (nFR nFR) schedules of reinforcement were used. Under these schedules, mouth‐contact responses on either operandum (spout) counted toward completion of both ratio requirements. Diagram 1 in the Supplementary Materials illustrates the fundamental difference between concurrent nonindependent nFR nFR schedules and the more familiar concurrent (independent) FR FR schedules. The upper frame shows that under commonly used FR FR schedules, responses on each manipulandum contribute independently toward completion of each schedule requirement. In contrast, the lower frame shows that with concurrent nonindependent nFR nFR schedules, responses contribute toward completion of both schedule requirements. When a ratio requirement was completed on the opposite side from which responding was occurring, the reinforcer delivery was withheld until the monkey responded on that spout. Upon collection of the reinforcer, the programmed ratio requirement was reset for the spout where the collection occurred. Table 2 shows some possible sequences of responses, schedule value changes, and deliveries under a concurrent nonindependent nFR 8 nFR 8 schedule.

TABLE 2.

POSSIBLE NFR RESPONSE SEQUENCE.

Spout 1 Spout 2
nFR 8 nFR 8
<< Response 1
nFR 7 nFR 7
<< Response 2
nFR 6 nFR 6
<< Response 3
nFR 5 nFR 5
<< Response 4
nFR 4 nFR 4
<< Response 5
nFR 3 nFR 3
<< Response 6
nFR 2 nFR 2
<< Response 7
nFR 1 nFR 1
<< Response 8
nFR 8 + Delivery Held
Response 9 >>
nFR 7 nFR 8 + Delivery
<< Response 10
nFR 6 nFR 7
<< Response 11
nFR 5 nFR 6
<< Response 12
nFR 4 nFR 5
<< Response 13
nFR 3 nFR 4
<< Response 14
nFR 2 nFR 3
<< Response 15
nFR 1 nFR 2
<< Response 16
nFR 8 + Delivery nFR 1
Response 17 >>
nFR 7 nFR 8 + Delivery
Response 18 >>
nFR 6 nFR 7
Response 19 >>
nFR 5 nFR 6
Response 20 >>
nFR 4 nFR 5
Response 21 >>
nFR 3 nFR 4
Response 22 >>
nFR 2 nFR 3
Response 23 >>
nFR 1 nFR 2
<< Response 24
nFR 8 + Delivery nFR 1
Response 25 >>
nFR 7 nFR 8 + Delivery
Open in a new tab

Note: << or > > indicates response on the corresponding spout.

The nFR size was increased for each monkey across blocks of six sessions of stable behavior until responding began to decline at a specific nFR value (which varied across subjects), at which point no further increases in the nFR value were made. All conditions were studied for at least six sessions, until behavior was stable. Stability was defined as the absence of appreciable increasing or decreasing trends in the number of responses per session of either available liquid across six consecutive sessions (Lemaire & Meisch, 1984).

After the session

The volume of liquid consumed from each reservoir was measured. One hour after the end of a session, the monkeys were fed.

Drugs

An alcohol stock solution (32% w/v) was made by dilutions of 95% v/v alcohol once a week. Alcohol concentrations were prepared by dilutions of the stock solution 20 hr before each session. Concentrated stock solutions of methadone hydrochloride (1.0 mg/mL) and cocaine hydrochloride (1.0 mg/mL), both from the National Institute on Drug Abuse, Rockville, MD, USA, were prepared once a week and refrigerated at 4°C. Daily drug solutions were mixed 2 hr before each session by adding appropriate amounts of water to a measured amount of stock solution. For monkey Crash, methadone functioned as a more effective reinforcer when placed in a 1% alcohol vehicle. The combination solution was therefore used rather than methadone alone. Drug concentrations are in terms of the salt. All drug solutions were at room temperature at the start of the session.

Data analysis

The mean numbers of responses, responses per liquid delivery, changeover responses, liquid deliveries, and volume consumed were calculated across the last six stable sessions of each condition. A standard error was computed for each mean.

RESULTS

Responses

As a function of ratio size, responses reinforced by delivery of both liquids initially increased and then decreased (Figure 1). Responses reinforced by deliveries of an 8% alcohol solution were much higher than those maintained by deliveries of the alternate drug solution. When points were redetermined in the descending series of ratio sizes, responses maintained by deliveries of the alcohol solution were higher when methadone solutions were the alternate reinforcer. The number of responses maintained by deliveries of the alternate liquids during the descending ratio series was frequently larger than that during the ascending ratio series (Figure 1).

FIGURE 1.

FIGURE 1

Open in a new tab

Alcohol responses were a curved inverted U‐shaped function of schedule size, whereas alternate liquid responses were far lower and less curved. The column on the left shows responses plotted on a linear scale, whereas the column on the right shows the same values on a log2 scale. Each point is a mean of six consecutive sessions of stable behavior; error bars indicate the SEM. Filled symbols: results from the ascending series of nFR sizes. Open symbols: results from the descending series. Note the different scales on the ordinates. A = alcohol; C = cocaine; M = methadone.

Responses per delivery

As schedule size increased, responses per delivery for both liquids increased (Figure 2). Alcohol‐reinforced responses were higher than responses reinforced by deliveries of the alternative liquid, and the differences between liquids became greater as the nFR value became larger.

FIGURE 2.

FIGURE 2

Open in a new tab

Alcohol responses per delivery were greater than alternate liquid responses per delivery, and the difference became greater as a function of schedule size. Each point is a mean of six consecutive sessions of stable behavior; error bars indicate the SEM. Filled symbols: results from the ascending series of nFR sizes. Open symbols: results from the descending series. Note the different scales on the ordinates.

Efficiency

Under the nFR schedules, responses on either operandum count toward completion of the response requirements on both sides. A consequence was that for a given number of reinforcers (deliveries of both liquids), fewer responses per liquid delivery occurred when responses were distributed across sides; that is, the distribution of responses toward each spout became more efficient. This measure of the efficiency of a distribution of responses between concurrent schedules, at each nFR schedule size, can be calculated by dividing the minimum possible number of responses required per reinforcer delivery by the total number of responses. The minimum number of responses per delivery equals the total number of deliveries obtained at a schedule value (alcohol plus alternate liquid) multiplied by one half the nFR value. This measure of efficiency increased as a direct function of the schedule size. It increased in the ascending series of schedule values and decreased with decreases in schedule size in the descending series of schedule values (Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

The efficiency quotient increased with increases in schedule size and decreased with decreases in schedule size. The ascending and descending functions were similar. Efficiency is defined as the minimum number of responses per ratio to obtain a delivery of alcohol and the alternate liquid, divided by total responses. The minimum is calculated by multiplying the sum of actual deliveries of each liquid by one half the ratio size. This minimum value is then divided by the total responses (alcohol plus alternate liquid). Each point is a mean of six consecutive sessions of stable behavior. Filled symbols: results from the ascending series of nFR sizes. Open symbols: results from the descending series.

Deliveries

Deliveries of alcohol decreased as a function of the nFR requirement (Figure 4). In contrast, in the ascending series of schedule sizes, deliveries of the alternate liquid first increased, and then at the highest schedule value declined. When schedule size was then decreased, deliveries of the alternative drug solution exceeded or equaled the initial values. These increases are different from what is observed with traditional FR FR schedules when the alternate reinforcer is not preferred, where deliveries of the alternate reinforcer generally remain low when schedule sizes are decreased. At the highest schedule requirement, the number of deliveries of both liquids converged.

FIGURE 4.

FIGURE 4

Open in a new tab

Alcohol deliveries directly decreased with the nFR requirement, whereas in contrast alternate liquid deliveries increased and then decreased. Values for the second determination of alcohol deliveries were higher than the initial series in 11 of 14 instances and the same for alternate liquid deliveries. Each point is a mean of six consecutive sessions of stable behavior; error bars indicate the SEM. Filled symbols: results from the ascending series of nFR sizes. Open symbols: results from the descending series. Note the different scale on the ordinate for Crash's alcohol‐methadone graph.

Percentage of deliveries of the alternate liquid

Deliveries of the alternate liquid as a percentage of all liquid deliveries increased directly with the schedule size (Figure 5). When the second determination was made in descending order, the percentage values for cocaine but not methadone were larger. The increasing percentage of total deliveries of the alternate liquid parallels the increase in differences in cost as schedule size increased (i.e., responses per delivery for 8% alcohol vs. the alternative liquid), as shown in Figure 2.

FIGURE 5.

FIGURE 5

Open in a new tab

The proportion of alternate liquid deliveries increased directly with nFR ratio size. When the nFR size was decreased, the values obtained with methadone were similar to initial values, whereas values obtained with cocaine were higher. Each point is a mean of six consecutive sessions of stable behavior. Filled symbols: results from the ascending series of nFR sizes. Open symbols: ROesults from the descending series.

Changeover responses

A consequence of changeover responses is a lowering of the number of responses required per liquid delivery. For the ascending series of nFR values, the total number of changeover responses increased and then decreased at the highest nFR size (Figure 6, left column). Often these responses followed liquid deliveries. When changeover responses per delivery were plotted, the resulting function increased with schedule size (Figure 6, right column).

FIGURE 6.

FIGURE 6

Open in a new tab

Changeover responses rose with the nFR value until the highest size was reached. Across monkeys in 13 of 14 pairs, the descending ratio results were higher than for the ascending series. When switches were adjusted per delivery, the relation with nFR size was direct. Each point is a mean of six consecutive sessions of stable behavior. Filled symbols: results from the ascending series of nFR sizes. Empty symbols: results from the descending series. Note the different scales on the ordinate.

Vehicle (water) deliveries

Upon completion of the series of increasing and decreasing nFR values, the water vehicle was substituted for the alternate solution and the increasing and decreasing series of ratio values was repeated with 8% alcohol concurrently available with water. Figure 7 shows that in most comparisons the vehicle values were below the alternate solution levels, especially during the descending nFR series. (Deliveries of the alternative‐drug solution are replotted from Figure 2.) Not shown are the deliveries of 8% alcohol that were concurrently scheduled. Thus, after the initial ratio sizes of 8 and 16, the alternate liquid frequently served as a reinforcer because consumption exceeded vehicle values.

FIGURE 7.

FIGURE 7

Open in a new tab

Water vehicle and alternate liquid deliveries across consecutive blocks of sessions. Eight percent alcohol was the concurrent liquid (alcohol values are not shown). In most comparisons water vehicle values were less than the drug values. For monkeys Crash and Lucas, water vehicle values were not obtained at nFR values of nFR 128 and 64 (Crash) and nFR 64 (Lucas) because responding was not maintained by 8% alcohol. Each point is a mean of six consecutive sessions of stable behavior; error bars indicate the SEM. Filled symbols: results for the alternate solution. Open symbols: results for the water vehicle.

Matching

Under concurrent nFR nFR or nVR nVR schedules, the relation between responses and deliveries is not fixed, and a matching relation for the ascending series of nFR values can be plotted as illustrated in Figure 8. The four panels of Figure 8 show that for the three monkeys there was a strong correspondence between relative deliveries and relative response rates, thus matching. The numerical value in front of x is the slope and is a measure of sensitivity to the reinforcer distribution; it ranged from .89 to .66. The R 2 value, a measure of the variance, was within a narrow range of .99 to .95.

FIGURE 8.

FIGURE 8

Open in a new tab

Performance under the nonindependent concurrent nFR nFR schedules. Mean response ratios in log units are plotted as a function of the mean delivery ratios in log units. Each point is the mean from the last six consecutive sessions of each condition for the ascending series of nFR values.

DISCUSSION

Under the nFR nFR schedules, monkeys regularly alternated between concurrently available drug solutions. These results contrast with findings obtained with traditional concurrent FR FR schedules, where the result is often the selection of only one of two concurrently available drugs (e.g., Seaman, Lordson, et al., 2022; Seaman, Rice, et al., 2022) or the unrelated consumption of two concurrently available drugs (Huskinson et al., 2015), unless the side positions of the drugs are reversed each session (Carroll, 1987). Concurrent FR FR schedules do not generate high rates of changeover responses because progress on one schedule does not change the probability of obtaining a reinforcer on the other schedule. In contrast, concurrent nFR nFR schedules promote choice, as responding on one schedule increases the probability of obtaining a reinforcer on the other schedule (MacDonall, 1988; Meisch & Spiga, 1998; Shull & Pliskoff, 1971). In the current study, concurrent nFR schedules were employed to induce intake of a nonpreferred drug instead of exclusive intake of a preferred drug that might be expected on traditional concurrent FR FR schedules. Use of the nonindependent ratio schedules provides the dependent variable of responses per delivery, or the amount paid for each delivery. The differences between liquids in responses per delivery at each schedule size are a measure of relative reinforcing effects. With traditional concurrent FR FR schedules, the number of responses per delivery is fixed and cannot reflect relative reinforcing magnitudes at each schedule value.

Responses per delivery is not a direct function of response rate and serves as a separate metric of relative reinforcing effects, which varies with schedule size. Additional measures can be derived, such as efficiency, which is calculated by dividing the minimum number of responses required per each reinforcer delivery by the total number of responses (Table 3). Changeover responses are important, for their occurrence is a measure of the relative reinforcing nature of concurrently available response consequences. Changeovers occur more often with nonindependent than with independent schedules (MacDonall, 1988; Meisch& Gomez, 2013; Shull & Pliskoff, 1971), at larger nonindependent schedule sizes (Meisch & Gomez, 2013), and with increases in the reinforcing effects of the concurrent reinforcer (Meisch et al., 2020). The present study extends findings with unequal reinforcers to conditions of equal nonindependent FR schedule sizes that varied across blocks of sessions. Choice of the less preferred drug reinforcer increased with increased schedule size. Thus, a determinant of polydrug use can be cost. Further on this matter, a commodity serves as a substitute when its consumption increases as the price of another commodity increases (Bickel et al., 1995; Hursh, 1993). In the present study the scheduled price for both reinforcers was held constant within sessions. However, the actual cost was not constant, for it was a function of the animal's response allocation. In the present study, as the cost of the preferred reinforcer increased, the relative cost of the nonpreferred reinforcer decreased and its consumption increased. Thus, the lesser reinforcer can be viewed as a substitute.

TABLE 3.

Calculation of efficiency.

Responses Responses Responses Deliveries Deliveries Deliveries Minimum Responses Efficiency m/t*100
0.4 mg/mL Cocaine 8% Alcohol Total 0.4 mg/mL Cocaine 8% Alcohol Total
nFR 8 20.67 2,168.17 2,188.83 3.50 275.00 278.50 1,114.00 50.9
nFR 16 48.17 3,915.50 3,963.67 6.33 246.83 253.17 2,025.33 51.1
nFR 32 311.33 6,177.50 6,488.83 20.33 200.83 221.17 3,538.67 54.5
nFR 64 549.50 5,779.00 6,328.50 29.17 98.00 127.17 4,069.33 64.3
nFR 128 398.83 4,802.67 5,201.50 19.17 40.50 59.67 3,818.67 73.4
nFR 64 1,322.50 5,242.00 6,564.50 42.33 100.50 142.83 4,570.67 69.6
nFR 32 984.67 4,922.83 5,907.50 52.33 173.17 225.50 3,608.00 61.1
nFR 16 1,063.50 4,172.67 5,236.17 92.33 283.33 375.67 3,005.33 57.4
nFR 8 428.67 2,581.00 3,009.67 73.83 339.67 413.50 1,654.00 55.0
Crash
Responses Responses Responses Deliveries Deliveries Deliveries
Methadone +1% Alcohol 8% Alcohol Total Methadone +1% Alcohol 8% Alcohol Total Minimum Responses Efficiency m/t*100
nFR 8 47.83 1,004.00 1,051.83 7.67 127.17 134.83 539.33 52.3
nFR 16 111.33 1,482.67 1,594.00 9.83 94.33 104.17 833.33 56.6
nFR 32 247.17 2,117.17 2,364.33 13.17 70.50 83.67 1,338.67 62.4
nFR 64 216.83 1,467.33 1,684.17 7.83 25.00 32.83 1,050.67 57.8
nFR 32 287.17 2,688.33 2,975.50 18.00 89.50 107.50 1,720.00 53.1
nFR 16 168.00 2,085.00 2,253.00 15.33 134.33 149.67 1,197.33 52.2
nFR 8 93.33 1,297.67 1,391.00 15.17 166.17 181.34 725.35 52.3
Lucas
Responses Responses Responses Deliveries Deliveries Deliveries Minimum Responses Efficiency m/t*100
0.2 Cocaine 8% Alcohol Total 0.2 Cocaine 8% Alcohol Total
nFR 8 285.7 2,628.3 338.0 52.3 336.7 389.0 1,556.0 0.59
nFR 16 484.2 4,150.5 4,634.7 49.7 268.8 318.5 2,548.0 0.55
nFR 32 1,383.8 4,261.5 5,645.3 89.3 168.5 257.8 4,125.3 0.73
nFR 64 2,319.8 4,987.3 7,307.2 86.0 111.2 197.2 6,309.3 0.86
nFR 128 1,026.2 2,301.5 3,327.7 21.5 25.3 46.8 2,997.3 0.90
nFR 64 2,335.0 4,131.8 6,466.8 76.3 91.0 167.3 5,354.7 0.83
nFR 32 3,031.0 4,564.2 7,595.2 176.0 219.0 395.0 6,320.0 0.83
nFR 16 1,065.3 4,125.0 5,190.3 104.8 284.8 389.7 3,117.3 0.60
nFR 8 1,329.2 2,093.7 3,422.8 185.3 277.5 462.8 1,851.3 0.54
Raja
Responses Responses Responses Deliveries Deliveries Deliveries Minimum Responses Efficiency m/t*100
0.4 M 8% Alcohol Total 0.4 M 8% Alcohol Total
nFR 8 142.5 1,608.2 1,750.7 26.0 203.5 229.5 918.0 0.52
nFR 16 314.5 2,866.0 3,180.5 42.5 188.7 231.2 1,849.3 0.58
nFR 32 667.3 3,139.7 3,807.0 51.3 115.5 166.8 2,669.3 0.70
nFR 64 461.8 1,687.7 2,149.5 26.0 33.0 59.0 1,888.0 0.88
nFR 32 427.8 3,553.0 3,980.8 47.0 121.3 168.3 2,693.3 0.68
nFR 16 245.5 3,497.7 3,743.2 40.3 225.8 266.2 2,129.3 0.57
nFR 8 278.3 2,462.2 2,740.5 53.2 312.5 365.7 1,462.7 0.53
Open in a new tab

Note: 0.4 Cocaine = 0.4 mg/mL cocaine; m = minimum; t = Total; M + 1% = 0.4 mg/mL methadone +1% alcohol; 0.4 M = 0.4 mg/mL methadone; 0.2 C = 0.2 mg/mL cocaine.

An important outcome of the present study was orderly intake of concurrently available drugs as a function of both experience and schedule value. Two general findings were that (1) as schedule size increased, intake of the preferred drug decreased, whereas intake of nonpreferred drug increased and (2) consumption of the nonpreferred drug remained elevated when schedule values were subsequently decreased. The ascending series of nFR values resulted in increased consumption of the drug solution that was concurrently present with 8% alcohol. The self‐administration experience obtained during the ascending series may have resulted in increased intake during the subsequent descending nFR series.

A problem in drug self‐administration studies is that a drug cannot come to function as a reinforcer unless it is self‐administered. However, low or zero rates of oral drug self‐administration occur in the absence of a preparatory history—upon initial exposure, reinforcing effects typically are absent. Thus, several methods have been employed to induce drug taking in the laboratory—for example, schedule‐induced polydipsia (Meisch & Thompson, 1971). In the present study, two factors interacted to generate oral intake of the less preferred drug—namely, the relative cost (responses per delivery of the less preferred drug relative to the number of responses per delivery of the preferred drug) and absolute cost (responses per delivery) of each drug. Thus, only one drug needs to be an effective reinforcer for interactions to occur. The present findings illustrate that schedule size (or price) is a determinant of choice between two commodities. This finding is reported in many studies of behavioral economics where the price of one commodity is held constant while the price of a second commodity is varied (Bickel et al., 2014; Carroll, 1993; Hursh, 1993). In contrast, in the present study at each programmed schedule size, the schedule size was always the same for both concurrently available drug solutions (commodities) and the number of responses per drug delivery became a dependent variable for each commodity. The subject could allocate how much was “paid” (the number of lip‐contact responses that were emitted) for deliveries of each of the concurrently available drug solutions. Thus, the design differs in a fundamental way from most prior behavioral economic studies of drug self‐administration. The number of deliveries was not a fixed function of number of responses. Therefore, a matching relation can be plotted, and this relation provides a link between matching and behavioral economics.

Concurrent nonindependent FR and VR schedules have been used in studies with pigeons and grain (Bellow & Lattal, 2023; Shull & Pliskoff, 1971), rats and food pellets (MacDonall, 1988), and rhesus monkeys and saccharin and drugs (Meisch & Gomez, 2013, 2016; Meisch et al., 2020; Meisch & Spiga, 1998). Results across studies are comparable. The data obtained with these schedules can be analyzed in several ways, including consumption as a function of price (behavioral economics) and log of relative response rates as a function of log of relative deliveries—that is, matching (Bellow & Lattal, 2023; MacDonall, 1988; Meisch & Spiga, 1998). Thus, use of these schedules provides an interface between behavioral economics and matching. Derived measures such as efficiency can be examined, and choice between qualitatively different reinforcers is feasible, as in the present study.

Less is known about concurrent nFR nFR schedules than about the more frequently used concurrent FR FR schedules. A limitation of the present study is that only one species was used: rhesus monkeys. Also, the range of drugs and doses was limited to three drugs and doses of 8% alcohol, 0.4 mg/mL methadone, and either 0.4 or 0.2 mg/mL cocaine. Thus, the potential generality of the findings needs to be strengthened by use of a broader range of conditions.

In summary, the current study illustrates that reinforcement schedule size (price) and relative cost can be determinants of acquisition and maintenance of polydrug use. Relative reinforcing effects of a nonpreferred drug can be increased by manipulations that alter the price of both concurrently available drugs. Responses per delivery provides a measure of relative reinforcing effects that is not directly related to response rates. In studies of choice between potential reinforcers, particularly behavioral economic studies, nonindependent ratio schedules result in the additional dependent variables of amount paid (responses/delivery) as well as relative amount paid per delivery (responses/delivery of one drug vs. the alternative drug). The functions generate possible baselines for the examination of candidate medications for polydrug abuse. The schedule arrangement results in multiple measures such as absolute and relative amounts paid per reinforcer delivery as well as a measure of efficiency. Higher rates of changeover responses occur that result in more frequent consumption of the less preferred reinforcer.

AUTHOR CONTRIBUTIONS

Richard Miesch (RM) conceived the study and designed and conducted the experiments, with assistance from Thomas Gomez (TG). Scott Lane (SL), TG and RM collaborated on data analysis and interpretation of results. RM wrote the initial draft of the manuscript, which was revised and edited together with TG and SL. All the authors approved the final version of the manuscript. All three authors stand accountable for all aspects of the work and its integrity.

CONFLICT OF INTEREST STATEMENT

None of the authors have a conflict of interest to disclose.

ETHICS APPROVAL

Animal care was in accordance with the guidelines of the Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council (2011), and the Institutional Animal Use and Care Committee of The University of Texas Health Science Center at Houston approved all procedures.

ACKNOWLEDGMENTS

We thank Gregory A. Lemaire for his helpful comments on the manuscript. This study was supported by funds received from the Center for Neurobehavioral Research on Addictions, and the University of Texas Health Science Center at Houston.

Meisch, R. A. , Gomez, T. H. , & Lane, S. D. (2025). Polydrug abuse: Choice between drugs as a function of concurrent nonindependent ratio sizes. Journal of the Experimental Analysis of Behavior, 124(2), e70054. 10.1002/jeab.70054

Editor‐in‐Chief: Suzanne H. Mitchell

Academic Editor: Erin Rasmussen.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

REFERENCES

  1. Amico, K. N. , Arnold, M. E. , Dourron, M. S. , Solomon, M. G. , & Schank, J. R. (2022). The effect of concurrent access to alcohol and oxycodone on self‐administration and reinstatement in rats. Psychopharmacology, 239(10), 3277–3286. 10.1007/s00213-022-06210-9 [DOI] [PubMed] [Google Scholar]
  2. Banks, M. L. , & Negus, S. S . (2012). Preclinical determinants of drug choice under concurrent schedules of drug self‐administration. Advances in Pharmacological Sciences, 2012, Article 281768. 10.1155/2012/281768 [DOI] [PMC free article] [PubMed]
  3. Bellow, D. , & Lattal, K. A. (2023). Choice dynamics in concurrent ratio schedules of reinforcement. Journal of the Experimental Analysis of Behavior, 119(2), 337–355. 10.1002/jeab.828 [DOI] [PubMed] [Google Scholar]
  4. Bickel, W. K. , DeGrandpre, R. J. , & Higgins, S. T. (1995). The behavioral economics of concurrent drug reinforcers: a review and reanalysis of drug self‐administration research. Psychopharmacology, 118(3), 250–259. 10.1007/BF02245952 [DOI] [PubMed] [Google Scholar]
  5. Bickel, W. K. , Koffarnus, M. N. , Moody, L. , & Wilson, A. G. (2014). The behavioral‐ and neuro‐economic process of temporal discounting: A candidate behavioral marker of addiction. Neuropharmacology, 76 Pt B(0 0), 518–527. 10.1016/j.neuropharm.2013.06.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cami, J. , & Ferré, M. (2003). Drug addiction. The New England Journal of Medicine, 349(10), 975–986. 10.1056/NEJMra023160 [DOI] [PubMed] [Google Scholar]
  7. Carr, K. D. (2002). Augmentation of drug reward by chronic food restriction: behavioral evidence and underlying mechanisms. Physiology & Behavior, 76(3), 353–364. 10.1016/s0031-9384(02)00759-x [DOI] [PubMed] [Google Scholar]
  8. Carroll, M. E. (1987). Self‐administration of orally‐delivered phencyclidine and ethanol under concurrent fixed‐ratio schedules in rhesus monkeys. Psychopharmacology, 93(1), 1–7. 10.1007/BF02439578 [DOI] [PubMed] [Google Scholar]
  9. Carroll, M. E. (1993). The economic context of drug and non‐drug reinforcers affects acquisition and maintenance of drug‐reinforced behavior and withdrawal effects. Drug and Alcohol Dependence, 33(2), 201–210. 10.1016/0376-8716(93)90061-t [DOI] [PubMed] [Google Scholar]
  10. Carroll, M. E. , & Meisch, R. A. (1984). Increased drug‐reinforced behavior due to food deprivation. In Thompson T., Dews P. B., and Barrett J. E. (Eds.), Advances in behavioral pharmacology (Vol. 4, pp. 47–88). Academic Press. 10.1016/B978-0-12-004704-8.50008-0 [DOI] [Google Scholar]
  11. Cheatle, M. D. (2015). Prescription opioid misuse, abuse, morbidity, and mortality: balancing effective pain management and safety. Pain Medicine, 1(Suppl 1), S3–S8. 10.1111/pme.12904 [DOI] [PubMed] [Google Scholar]
  12. Comer, S. D. , Metz, V. E. , Cooper, Z. D. , Kowalczyk, W. J. , Jones, J. D. , Sullivan, M. A. , Manubay, J. M. , Vosburg, S. K. , Smith, M. E. , Peyser, D. , & Saccone, P. A. (2013). Comparison of a drug versus money and drug versus drug self‐administration choice procedure with oxycodone and morphine in opioid addicts. Behavioural Pharmacology, 24(5–6), 504–516. 10.1097/FBP.0b013e328363d1c4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Connor, J. P. , Gullo, M. J. , White, A. , & Kelly, A. B. (2014). Polysubstance use: diagnostic challenges, patterns of use and health. Current Opinion in Psychiatry, 27(4), 269–275. 10.1097/YCO.0000000000000069 [DOI] [PubMed] [Google Scholar]
  14. Gieske, D. (1978). Integrated drinking device for monkeys (Tech. Rep. PR‐78‐1). University of Minnesota, Department of Psychiatry. [Google Scholar]
  15. Haney, M. , Rubin, E. , & Foltin, R. W. (2011). Aripiprazole maintenance increases smoked cocaine self‐administration in humans. Psychopharmacology, 216(3), 379–387. 10.1007/s00213-011-2231-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Harper, C. E. , Mata, D. C. , & Lee, D. (2023). The impact of fentanyl on DUIDs and traffic fatalities: Blood and oral fluid data. Journal of Forensic Sciences, 68(5), 1686–1697. 10.1111/1556-4029.15334 [DOI] [PubMed] [Google Scholar]
  17. Henningfield, J. E. , & Meisch, R. A. (1976). Drinking device for rhesus monkeys. Pharmacology, Biochemistry, and Behavior, 4(5), 609–610. 10.1016/0091-3057(76)90204-5 [DOI] [PubMed] [Google Scholar]
  18. Henningfield, J. E. , & Meisch, R. A. (1978). Ethanol drinking by rhesus monkeys as a function of concentration. Psychopharmacology, 57(2), 133–136. 10.1007/BF00426877 [DOI] [PubMed] [Google Scholar]
  19. Hursh, S. R. (1993). Behavioral economics of drug self‐administration: An introduction. Drug and Alcohol Dependence, 33(2), 165–172. 10.1016/0376-8716(93)90058-x [DOI] [PubMed] [Google Scholar]
  20. Huskinson, S. L. , Freeman, K. B. , & Woolverton, W. L. (2015). Self‐administration of cocaine and remifentanil by monkeys under concurrent‐access conditions. Psychopharmacology, 232(2), 321–330. 10.1007/s00213-014-3661-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Iglauer, C. , & Woods, J. H. (1974). Concurrent performances: reinforcement by different doses of intravenous cocaine in rhesus monkeys. Journal of the Experimental Analysis of Behavior, 22(1), 179–196. 10.1901/jeab.1974.22-179 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Janssen, E. , & Vuolo, M. (2024). Correlates of stimulant use among people who use heroin undergoing treatment in out‐patient facilities in France, 2010–2020. Substance Use & Misuse, 59(3), 353–361. 10.1080/10826084.2023.2270682 [DOI] [PubMed] [Google Scholar]
  23. Kedia, S. , Sell, M. A. , & Relyea, G. (2007). Mono‐ versus polydrug abuse patterns among publicly funded clients. Substance Abuse Treatment, Prevention, and Policy, 2, Article 33. 10.1186/1747-597X-2-33 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kemnitz J. W. (2011). Calorie restriction and aging in nonhuman primates. ILAR Journal, 52(1), 66–77. 10.1093/ilar.52.1.66 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kliner, D. J. , & Meisch, R. A. (1989). Oral pentobarbital intake in rhesus monkeys: Effects of drug concentration under conditions of food deprivation and satiation. Pharmacology, Biochemistry, and Behavior, 32(1), 347–354. 10.1016/0091-3057(89)90253-0 [DOI] [PubMed] [Google Scholar]
  26. Lamb, R. J. , Maguire, D. R. , Ginsburg, B. C. , Pinkston, J. W. , & France, C. P. (2016). Determinants of choice, and vulnerability and recovery in addiction. Behavioural Processes, 127, 35–42. 10.1016/j.beproc.2016.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lemaire, G. A. , & Meisch, R. A. (1984). Pentobarbital self‐administration in rhesus monkeys: drug concentration and fixed‐ratio size interactions. Journal of the Experimental Analysis of Behavior, 42(1), 37–49. 10.1901/jeab.1984.42-37 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. MacDonall, J. S. (1988). Concurrent variable‐ratio schedules: Implications for the generalized matching law. Journal of the Experimental Analysis of Behavior, 50(1), 55–64. 10.1901/jeab.1988.50-55 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Macenski, M. J. , & Meisch, R. A. (1992). Ethanol‐reinforced responding of naive rhesus monkeys: acquisition without induction procedures. Alcohol, 9(6), 547–554. 10.1016/0741-8329(92)90095-r [DOI] [PubMed] [Google Scholar]
  30. Mattison, J. A. , Colman, R. J. , Beasley, T. M. , Allison, D. B. , Kemnitz, J. W. , Roth, G. S. , Ingram, D. K. , Weindruch, R. , de Cabo, R. , & Anderson, R. M. (2017). Caloric restriction improves health and survival of rhesus monkeys. Nature Communications, 8, Article 14063. 10.1038/ncomms14063 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Meisch, R. A. , & Gomez, T. H. (2013). Drug self‐administration studies: a novel reinforcement schedule enhances choice. Behavioural Pharmacology, 24(3), 155–163. 10.1097/FBP.0b013e32836104cd [DOI] [PubMed] [Google Scholar]
  32. Meisch, R. A. , & Gomez, T. H. (2016). Concurrent nonindependent fixed‐ratio schedules of alcohol self‐administration: Effects of schedule size on choice. Journal of the Experimental Analysis of Behavior, 106(1), 75–92. 10.1002/jeab.215 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Meisch, R. A. , Gomez, T. H. , & Lane, S. D. (2020). Equal response rates maintained by concurrent drug and nondrug reinforcers: a design for treatment evaluation. Behavioural Pharmacology, 31(5), 458–464. 10.1097/FBP.0000000000000525 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Meisch, R. A. , & Lemaire, G. A. (1988). Oral self‐administration of pentobarbital by rhesus monkeys: Relative reinforcing effects under concurrent fixed‐ratio schedules. Journal of the Experimental Analysis of Behavior, 50(1), 75–86. 10.1901/jeab.1988.50-75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Meisch, R. A. , & Lemaire, G. A. (1990). Reinforcing effects of a pentobarbital‐ethanol combination relative to each drug alone. Pharmacology, Biochemistry, and Behavior, 35(2), 443–450. 10.1016/0091-3057(90)90182-h [DOI] [PubMed] [Google Scholar]
  36. Meisch, R. A. , & Lemaire, G. A. (1991). Effects of feeding condition on orally delivered ethanol as a reinforcer for rhesus monkeys. Alcohol, 8(1), 55–63. 10.1016/0741-8329(91)91264-3 [DOI] [PubMed] [Google Scholar]
  37. Meisch, R. A. , & Spiga, R. (1998). Matching under nonindependent variable‐ratio schedules of drug reinforcement. Journal of the Experimental Analysis of Behavior, 70(1), 23–34. 10.1901/jeab.1998.70-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Meisch, R. A. , & Stewart, R. B. (1995). Relative reinforcing effects of different doses of orally delivered cocaine. Drug and Alcohol Dependence, 37(2), 141–147. 10.1016/0376-8716(94)01068-v [DOI] [PubMed] [Google Scholar]
  39. Meisch, R. A. , Stewart, R. B. , & Wang, N. S. (1996). Orally delivered methadone as a reinforcer for rhesus monkeys: The relationship between drug concentration and choice. Pharmacology, Biochemistry, and Behavior, 54(3), 547–554. 10.1016/0091-3057(95)02214-7 [DOI] [PubMed] [Google Scholar]
  40. Meisch, R. A. , & Thompson, T. (1971). Ethanol intake in the absence of concurrent food reinforcement. Psychopharmacologia, 22(1), 72–79. 10.1007/BF00401469 [DOI] [PubMed] [Google Scholar]
  41. National Research Council (US) Committee for the Update of the Guide for the Care and Use of Laboratory Animals . (2011). In Guide for the Care and Use of Laboratory Animals . (8th ed., pp 6–9). National Academies Press. [PubMed]
  42. Negus S. S. (2005). Interactions between the reinforcing effects of cocaine and heroin in a drug‐vs‐food choice procedure in rhesus monkeys: A dose‐addition analysis. Psychopharmacology, 180(1), 115–124. 10.1007/s00213-004-2133-y [DOI] [PubMed] [Google Scholar]
  43. Potoukian, R. B. , Gonyea, J. , Shoff, E. N. , Hime, G. W. , & Moore, D. M. (2023). Prevalence of xylazine in overdose cases: An analysis of Miami‐Dade County medical examiner case data. Journal of Forensic Sciences, 68(6), 2205–2210. 10.1111/1556-4029.15375 [DOI] [PubMed] [Google Scholar]
  44. Pugh, T. D. , Klopp, R. G. , & Weindruch, R. (1999). Controlling caloric consumption: protocols for rodents and rhesus monkeys. Neurobiology of Aging, 20(2), 157–165. 10.1016/s0197-4580(99)00043-3 [DOI] [PubMed] [Google Scholar]
  45. Seaman, R. W., Jr. , Lordson, C. , & Collins, G. T. (2022). Modeling stimulant and opioid co‐use in rats provided concurrent access to methamphetamine and fentanyl. Frontiers in Psychiatry, 13, Article 814574. 10.3389/fpsyt.2022.814574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Seaman, R. W., Jr. , Rice, K. C. , & Collins, G. T. (2022). Relative reinforcing effects of cocaine and 3,4‐methylenedioxypyrovalerone (MDPV) under a concurrent access self‐administration procedure in rats. Drug and Alcohol Dependence, 232, Article 109299. 10.1016/j.drugalcdep.2022.109299 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shull, R. L. , & Pliskoff, S. S. (1971). Changeover behavior under pairs of fixed‐ratio and variable‐ratio schedules of reinforcement. Journal of the Experimental Analysis of Behavior, 16(1), 75–79. 10.1901/jeab.1971.16-75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Stewart, R. B. , Bass, A. A. , Wang, N. S. , & Meisch, R. A. (1996). Ethanol as an oral reinforcer in normal weight rhesus monkeys: Dose‐response functions. Alcohol, 13(4), 341–346. 10.1016/0741-8329(96)00004-3 [DOI] [PubMed] [Google Scholar]
  49. Walsh, S. L. , Nuzzo, P. A. , Babalonis, S. , Casselton, V. , & Lofwall, M. R. (2016). Intranasal buprenorphine alone and in combination with naloxone: Abuse liability and reinforcing efficacy in physically dependent opioid abusers. Drug and Alcohol Dependence, 162, 190–198. 10.1016/j.drugalcdep.2016.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wang, N. S. , Brown, V. L. , Grabowski, J. , & Meisch, R. A. (2001). Reinforcement by orally delivered methadone, cocaine, and methadone‐cocaine combinations in rhesus monkeys: are the combinations better reinforcers? Psychopharmacology, 156(1), 63–72. 10.1007/s002130100731 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Journal of the Experimental Analysis of Behavior are provided here courtesy of Wiley

RESOURCES