Effects Of Amphetamine-Cns Depressant Combinations And Of Other Cns Stimulants In Four-Choice Drug Discriminations

Abstract

Three pigeons were trained to discriminate among 5 mg/kg pentobarbital, 2 mg/kg amphetamine, a combination of these two drugs at these doses, and saline using a four-choice procedure (amphetamine–pentobarbital group). Three other pigeons were trained to discriminate among 5 mg/kg morphine, 2 mg/kg methamphetamine, a combination of these two drugs at these doses, and saline (methamphetamine–morphine group). After 10 to 13 months of training, the pigeons averaged more than 90% of their responses on the appropriate key during training sessions. In subsequent testing, dose-response curves were determined for the individual drugs, for a wide range of dose combinations of the training drugs, and for two drugs to which the pigeons had not been exposed previously (pseudoephedrine and nicotine). After low test doses of the training drugs, pigeons responded on the saline key. As the dose increased, responding on the key associated with that drug during training sessions increased. When training drugs were combined at doses that were not discriminable when given alone, responding occurred on the saline key. When a discriminable dose of one training drug was combined with a nondiscriminable dose of the other training drug, responding occurred on the key associated with the discriminable dose. When both drugs were given at discriminable doses, responding almost always occurred on the drug-combination key. The response-rate decreasing effects of pentobarbital and amphetamine were mutually antagonized when the drugs were combined, but the rate-decreasing effects of morphine and methamphetamine were not. After low doses of pseudoephedrine and nicotine, pigeons in both groups responded on the saline key. After higher doses of pseudoephedrine and nicotine, responding in the amphetamine–pentobarbital group occurred primarily on the amphetamine key. In the methamphetamine–morphine group, higher doses of pseudoephedrine and especially nicotine engendered more responding on the combination key than had occurred in the other group. The four-choice procedure can reveal subtle effects in the discrimination of individual drugs and drug combinations that are not apparent with procedures offering fewer response alternatives.

Drug-discrimination procedures have been widely used to study the effects of abused drugs, but there have been few studies on the discriminative-stimulus effects of drug combinations despite widespread polydrug abuse. Most investigators have used a two-choice procedure to study the effects of individual drugs. Under the two-choice procedure, responses on one manipulandum are reinforced if a training drug was administered before the session, and responses on a second manipulandum are reinforced if the drug vehicle was administered before the session. Once the discrimination is well established, testing of other doses of the training drugs or of other drugs can be conducted to determine if they produce discriminative stimuli that substitute for those of the training dose.

There are several options for applying the two-choice procedure to studying the discriminative stimulus effects of drug combinations. Animals have been trained to discriminate between a drug combination and vehicle (Mariathasan & Stolerman, 1994; Stolerman, Mariathasan, & White, 1999), or between the drug combination and either of the component drugs presented alone (Mariathasan & Stolerman, 1994; Mariathasan, Stolerman & White, 1999a, 1999b; Stolerman, Mariathasan, & White, 1999). Most studies on the discriminative stimulus effects of drug combinations have used the procedure with animals trained to discriminate between a drug combination and saline. When animals that have been trained under this procedure are tested with discriminable doses of the individual drugs, responding is usually restricted to the combination key (Harrison, Jenkins, Rocha, Lytle, Jung, & Oglesby, 1998; Mariathasan & Stolerman, 1992, 1994; Negus, Gatch, & Mello, 1998; Shoaib, Baumann, Rothman, Goldberg, & Schindler, 1997; Stolerman, Mariathasan, White, & Olufsen, 1999). In a few cases, the generalization has been shown to be only partial (,Mariathasan, Stolerman, & White, 1999a, 1999b), and in rare instances, absent (Carlezon, Kosten, & Nestler, 2000). When animals trained to discriminate between a drug combination and the component drugs are tested with doses of the individual component drugs, less responding occurs on the combination key (Mariathasan & Stolerman, 1994; Stolerman, Mariathasan, & White, 1999).

Both of these procedures have limitations. When animals are trained to discriminate between a drug combination and vehicle, the procedure does not easily differentiate between the individual drugs in the combination, nor can it differentiate between vehicle and drug stimuli that are qualitatively different from the stimuli produced by the combination or by the individual drugs in the combination. In the former case, either of the individual drugs of the combination usually produces responding on the drug key; in the latter case, drugs with stimulus properties that differ from those of the component drugs used in the training combination usually produce responding on the vehicle key. Although training animals to discriminate between drug combinations and the individual drugs in the combination appears to have greater specificity in differentiating among drug stimuli, there is no appropriate response choice available if a drug does not produce a discriminative stimulus, nor is there an alternative response available if the drug produces qualitatively different discriminative stimuli from those produced by the drug combination or by either of the individual drugs.

McMillan and Li (2002) argued that there are four stimulus conditions of importance when studying the discriminative stimuli produced by drug combinations: the discriminative stimuli produced by the drug combination, the discriminative stimuli produced by each of the components of the combination when administered alone, and the absence of all three of these discriminative stimuli. They suggested that by using a four-choice drug discrimination procedure, all of these states could be studied in a single experiment. Toward this end, pigeons were trained to discriminate among pentobarbital, morphine, a combination of these two drugs, and saline. Low doses of pentobarbital produced responding on the saline key and higher doses produced responding on the pentobarbital key. Low doses of morphine produced responding on the saline key and higher doses produced responding on the morphine key. All doses of pentobarbital and morphine that were discriminated when given alone produced responding on the combination key when given in combination. Methamphetamine, a drug with discriminative stimuli different from those of pentobarbital and morphine, produced responding on the saline key. Thus the four-choice procedure differentiated among the discriminative stimuli produced by the combination and by each of the components of the combination, as well as provided an option for responding when none of these discriminative stimuli were present.

The first purpose of the present experiments was to determine the generality of the observations of McMillan and Li (2002) by studying some new drug combinations. McMillan and Li studied the discriminative-stimulus properties of two depressant drugs (pentobarbital and morphine), alone and in combination, using the four-choice procedure. The current experiments focused on stimulant-depressant interactions using the four-choice procedure with one group of pigeons trained to discriminate among pentobarbital, amphetamine, a combination of these drugs, and saline, and a second group trained to discriminate among morphine, methamphetamine, a combination of these drugs, and saline.

A second purpose of the present experiments was to determine if cumulative-dosing procedures could be used to study these complex drug interactions under the four-key drug-discrimination procedure. To study drug combinations across a range of doses requires a large number of experiments. For example, McMillan and Li (2002) established four-point dose-response curves for each of the training drugs given alone and then studied all possible combinations of the four doses of each drug. This required 24 separate test sessions to make single observations of each dose combination in each subject. If a cumulative-dosing procedure could be applied successfully, it would eliminate 75% of these test sessions. Cumulative-dosing procedures have been used previously in drug-discrimination experiments (Bertalmio, Herling, Hampton, Winger, & Woods, 1982; McMillan, Cole-Fullenwider, Hardwick, & Wenger, 1982; Wenger, 1980), but their use has not been reported for four-choice drug discrimination.

The third purpose of this study was to use the four-choice procedure to determine the effects of two drugs, pseudoephedrine and nicotine, to which the pigeons had not been exposed previously. Both compounds are naturally occurring plant alkaloids with central nervous system (CNS) stimulant properties. Pseudoephedrine is an isomer of ephedrine, a compound with four isomers due to its two asymmetric carbons. Ephedrines are structurally similar to methamphetamine, differing only by a hydroxyl group substitution for hydrogen at the β-carbon position (Young, Bondarev, & Glennon, 1999). Although (−)- and racemic ephedrine have been studied in pigeons trained to discriminate amphetamine from saline (Ercil & France, 2003), we could find no cases in which the pseudoephedrine isomers have been studied in pigeons. In rats trained to discriminate amphetamine from saline, most subjects responded on the amphetamine lever following pseudoephedrine at doses of 20 and 40 mg/kg, i.p. (Tongjaroenbuangam, Meksuriyen, Govitrapong, Kotchabhakdi, & Baldwin, 1998). In contrast, Young and Glennon (2000) reported that 15 mg/kg pseudoephedrine produced 61% amphetamine-appropriate responding, although only at doses that greatly decreased response rates. In rhesus monkeys trained to discriminate amphetamine administered by the intragastric route, pseudoephedrine substituted for amphetamine in 2 of 3 subjects (Anderson, Winger, Woods, & Woolverton, 2001).

The naturally occurring isomer of nicotine in tobacco, and the one used here, is (−)nicotine (Bowman & Rand, 1980). Few studies have studied the discriminative-stimulus properties of nicotine in pigeons. In pigeons trained to make a three-choice discrimination among pentobarbital, amphetamine, and saline, low doses of nicotine resulted in saline responding, but higher doses resulted in dose-dependent but partial (<80%) responding on the amphetamine-appropriate key. After retraining these pigeons also to discriminate morphine on the pentobarbital key and caffeine on the amphetamine key, the same dose range of nicotine produced responding predominantly on the saline key (Li & McMillan, 2003). In rats trained to discriminate amphetamine from saline, nicotine also resulted in partial substitution for amphetamine (Bardo, Bevins, Klebaur, Crooks, & Dwoskin, 1997). The study of pseudoephedrine and nicotine in drug discrimination procedures in which amphetamine and methamphetamine serve as the training drugs but where more response options are available might clarify some of these ambiguities.

METHOD

Subjects

Six adult male White Carneau pigeons (Palmetto Pigeon Plant, Sumter, SC) served as subjects. Three pigeons (see below) had served in previous experiments, and the other 3 pigeons were experimentally naive at the beginning of these experiments. Pigeons were housed individually with free access to water and grit in a temperature- and humidity-controlled room that was maintained under a 12:12 hr light/dark cycle. During the study, the pigeons were maintained at 80 to 85% of their free-feeding weights (421 g to 515 g), by food earned during the experimental sessions and from postsession supplemental feeding (Purina® Pigeon Chow Checkers 5405, Purina Mills, LLC, St. Louis, MO). Procedures used during these experiments were in accord with the Institutional Animal Care and Use Committee of the University of Arkansas for Medical Sciences.

Apparatus

The test chamber was a Gerbrands Model G7313 standard pigeon test cage, measuring 31 cm wide by 31 cm long by 29 cm high, enclosed in a Gerbrands Model G7211 sound- and light-attenuating enclosure. Two 28-V DC lights mounted near the ceiling illuminated the test chamber except during food presentations, when only a light inside the food hopper was illuminated. The front panel of the test cage was modified so that four pigeon response keys (Gerbrands Model G6315), each 2.0 cm in diameter, were mounted 3 cm apart in a row 20 cm above the grid floor. When operative, the left key was red, the left-center key was white, the right-center key was green, and the right key was blue. Centered between the middle two response keys at floor level was a food hopper in which pigeon chow could be presented when schedule contingencies were met. A desktop PC computer programmed in MedState Notation^TM using MED-PC® software controlled the experimental contingencies and recorded the data through an interface (MED Associates).

Procedure

Pigeons P360, P347, and P380 had been trained in previous drug-discrimination experiments to discriminate between pentobarbital and saline (McMillan, Hardwick, & Li, 2001; McMillan & Li, 2000). These pigeons were further trained to discriminate among pentobarbital, amphetamine, a combination of these drugs, and saline (the amphetamine–pentobarbital group). The remaining 3 pigeons, P404, P411, and P412, had not been used previously and were trained to discriminate among morphine, methamphetamine, a combination of these drugs, and saline (the methamphetamine–morphine group).

For the amphetamine–pentobarbital pigeons with previous experience discriminating 5 mg/kg pentobarbital from saline under a two-choice procedure, a third key was introduced and responding on this key was reinforced only if 2 mg/kg amphetamine had been administered before the session. After responding had stabilized for this three-choice drug discrimination, a fourth key was added, and responding was reinforced on this key only when a combination of 5 mg/kg pentobarbital and 2 mg/kg amphetamine had been administered before the session. The reinforcement schedule under which food was presented for responding on the correct key during training was a fixed-ratio (FR) 20.

The pigeons in the methamphetamine–morphine group were trained to key peck on a single key for food, after which a second key was introduced, and responses on that key were reinforced only if 5 mg/kg morphine had been administered before the session. If saline was administered before the session, then only responses on the first key were reinforced. After the discrimination between morphine and saline was acquired, a third key was lighted and responses on that key were reinforced only if 2 mg/kg methamphetamine had been administered before the session. Finally, a fourth key was introduced and responses on that key were reinforced only if both 5 mg/kg morphine and 2 mg/kg methamphetamine had been administered before the session. Under all conditions of the experiment, key-peck responses on the appropriate key were reinforced under an FR 20 schedule by operating the food hopper to permit 4-s access to food.

Training sessions ended after 40 min or after 20 reinforcer presentations under the FR 20 schedule, whichever occurred first. Training sessions were conducted 5 or 6 days a week with at least one session under each stimulus condition each week. Training continued until performance showed no further signs of improvement for 1 month. The final level of performance for individual pigeons under each training condition is shown in Table 1.

Table 1. The percentage of total responses that occurred on each response key averaged across six training sessions that occurred prior to determining the dose-response curves for individual drugs and drug combinations. Standard errors are shown in parentheses.

Injection	Pigeon	Response key
		Saline	Pentobarbital	Amphetamine	Combined
Saline	P347	93.6 (3.1)	0.0 (0.0)	2.3 (1.0)	4.2 (2.8)
	P260	81.9 (3.7)	5.6 (5.1)	2.3 (4.1)	0.0 (0.0)
	P380	89.9 (2.6)	4.4 (2.2)	3.6 (1.7)	2.1 (1.0)
Pentobarbital	P347	0.0 (0.0)	94.9 (2.6)	0.8 (0.8)	4.3 (2.8)
	P260	2.9 (2.1)	84.8 (11.2)	12.3 (11.5)	0.0 (0.0)
	P380	11.6 (4.4)	75.7 (8.5)	2.5 (2.5)	10.3 (10.3)
Amphetamine	P347	0.8 (0.8)	0.0 (0.0)	98.4 (1.0)	0.8 (0.8)
	P260	6.7 (3.6)	3.6 (1.8)	89.7 (3.8)	0.0 (0.0)
	P380	4.6 (4.0)	0.0 (0.0)	90.1 (6.3)	5.3 (4.1)
Combined	P347	0.0 (0.0)	0.0 (0.0)	0.8 (1.9)	99.2 (0.8)
	P260	3.9 (3.3)	0.0 (0.0)	6.9 (4.5)	89.1 (5.1)
	P380	6.0 (3.4)	1.5 (0.9)	2.0 (1.3)	90.7 (4.4)

		Saline	Morphine	Methamphetamine	Combined
Saline	P404	95.7 (2.9)	1.5 (1.5)	0.0 (0.0)	2.8 (2.8)
	P411	98.4 (1.0)	0.8 (1.9)	0.8 (1.9)	0.0 (0.0)
	P412	92.4 (9.1)	2.8 (2.8)	4.9 (3.4)	0.0 (0.0)
Morphine	P404	0.8 (0.8)	95.9 (3.3)	0.0 (0.0)	3.3 (3.3)
	P411	2.9 (2.1)	84.8 (11.2)	12.3 (11.5)	0.0 (0.0)
	P412	4.4 (2.3)	92.0 (2.6)	0.0 (0.0)	3.6 (2.4)
Methamphetamine	P404	3.0 (2.1)	0.0 (0.0)	97.0 (2.2)	0.0 (0.0)
	P411	6.7 (3.6)	3.6 (1.8)	89.7 (3.8)	0.0 (0.0)
	P412	3.0 (2.2)	0.0 (0.0)	93.2 (3.9)	3.8 (3.8)
Combined	P404	0.0 (0.0)	5.9 (2.6)	0.0 (0.0)	94.1 (2.6)
	P411	1.6 (1.0)	0.0 (0.0)	3.3 (3.3)	95.1 (3.2)
	P412	0.8 (0.8)	0.2 (0.2)	0.8 (0.8)	97.7 (1.6)

During subsequent test sessions, individual doses of pentobarbital or amphetamine were administered to the pigeons in the pentobarbital–amphetamine group, and individual doses of morphine or methamphetamine were administered to the pigeons in the methamphetamine–morphine group to determine dose-response relations. The procedure for test sessions was the same as for the training sessions, except that the test session ended after 20 responses had been made on any key and the food reinforcer was presented, or after 15 min had elapsed, whichever occurred first. These drug substitution tests were conducted 2 days a week, on Tuesday and on either Friday or Saturday, with training sessions continuing on the intervening days.

After completion of these dose-response curves, various combinations of doses of pentobarbital and amphetamine were studied with the amphetamine–pentobarbital group and various combinations of doses of morphine and methamphetamine were studied with the methamphetamine–morphine group under procedures similar to those employed for determining the dose-response curves for the individual drugs.

Upon completion of these dose-response curves, a series of experiments using cumulative dosing was performed. During this period, training sessions continued as usual; however, when drug substitution tests were performed, doses were studied cumulatively so that an entire dose-response curve could be determined during a single session. For the pigeons in the amphetamine–pentobarbital group, cumulative dose-response curves were determined for pentobarbital, amphetamine, pseudoephedrine, nicotine, and combinations of 1.0 mg/kg amphetamine plus different doses of pentobarbital. For the pigeons in the methamphetamine–morphine group, cumulative dose-response curves were determined for morphine, methamphetamine, pseudoephedrine, nicotine, and combinations of 1 mg/kg methamphetamine with different doses of morphine. Under the cumulative-dosing procedure, the lowest dose of a drug was given and the pigeon was placed in the darkened chamber. Ten minutes later a test trial began. After the delivery of one reinforcer or 5 min, whichever occurred first, the trial ended and the houselights were extinguished. The pigeon was then removed from the chamber and administered a second dose, 10 min after which a second trial would ensue. This procedure was repeated until the cumulative dose-response curve had been determined, usually three or four doses. For cumulative dose-response curves, the doses shown represent the sum of all doses given during the session. These sessions will be referred to as cumulative-dose sessions to differentiate them from the single-dose sessions in which only one dose or one drug combination was given.

Data Analysis

The percentage of responses on each key was calculated by dividing the number of responses on each key by the total number of responses on all four keys and converting to a percentage. Response keys are identified by the training drug treatments. The key on which the pigeons responded most frequently is referred to as the key on which they responded “predominantly.” The total number of responses on all keys was divided by the session time to calculate the overall rate of responding during the session.

Drugs

The drugs used were purchased from commercial sources. The drugs were pentobarbital sodium, (+)methamphetamine hydrochloride, (+)amphetamine sulfate, (−)morphine sulfate, (+)pseudoephedrine hydrochloride, and (−)nicotine hydrogen tartrate. Doses are expressed as the salts and were administered as intramuscular injections 10 min before the session in a volume of 1.0 ml/kg of body weight. After injection, the pigeons were placed in the darkened test chamber until the session began. When drug combinations were studied, one drug was given into one side of the breast muscle, followed by injection of the second drug into the opposite side of the breast muscle. During cumulative dosing, a single dose of pentobarbital or morphine was given, followed by increasing doses of amphetamine or methamphetamine given with alternation of successive doses between breast muscles.

RESULTS

Pigeons in the amphetamine–pentobarbital group (P347, P260, and P380) required 10 additional months of training to discriminate among pentobarbital, amphetamine, these two drugs in combination, and saline. Similarly, the drug-naive pigeons in the methamphetamine–morphine group (P404, P411, and P412) required 13 months of training before their behavior stabilized and they were able to discriminate among morphine, methamphetamine, the combination, and saline. Table 1 shows the level of performance that each subject reached after this extensive training. The amphetamine–pentobarbital group averaged 88.5% of their responses on the saline key after saline, 85.1% of their responses on the pentobarbital key after 5 mg/kg pentobarbital, 92.7% of their responses on the amphetamine key after 2 mg/kg amphetamine, and 93.0% of their responses on the combination key after being administered 5.0 mg/kg pentobarbital plus 2.0 mg/kg amphetamine. The methamphetamine–morphine group averaged 95.5% of their responses on the saline key after saline, 90.9% of responses on the morphine key after 5 mg/kg morphine, 93.3% of responses on the methamphetamine key after 2 mg/kg methamphetamine, and 95.7% of responses on the combination key after 5.0 mg/kg morphine plus 2.0 mg/kg methamphetamine.

Figure 1 shows the pattern of responding on each key following increasing doses of pentobarbital and amphetamine for each pigeon in the amphetamine–pentobarbital group. The first row of Figure 1 shows the pentobarbital dose-response curves determined using single-dose test sessions, whereas the second row shows the pentobarbital dose-response curves determined using cumulative-dose procedures. All pigeons responded predominately on the saline-appropriate key after the lowest dose of pentobarbital (1.0 mg/kg). As the dose of pentobarbital increased, responding shifted primarily to the pentobarbital-appropriate key. Although there were slight differences among pigeons in the doses at which the switch from the saline key to the pentobarbital key occurred, there were no systematic differences between single-dose and cumulative-dose testing procedures. The final two rows of Figure 1 show the dose-response curves determined in these same subjects for amphetamine using single-dose procedures (third row) compared to cumulative-dose procedures (bottom row). As was seen with pentobarbital, low doses of amphetamine (0.3 mg/kg, and to some extent 1.0 mg/kg) resulted in responding predominantly on the saline key. As the dose of amphetamine increased, responding shifted to the amphetamine key. Again, there were no consistent differences between cumulative-dose and single-dose testing procedures.

Fig. 1. Patterns of responding on each key following increasing doses of pentobarbital and amphetamine in pigeons trained to discriminate among saline, pentobarbital, amphetamine, and a combination of pentobarbital and amphetamine (amphetamine–pentobarbital group).

The top row shows the pentobarbital dose-response curve determined using single-dose procedures and the second row shows the pentobarbital dose-response curve determined using cumulative-dose procedures. The bottom two rows show amphetamine dose-response curves determined using single-dose procedures (third row) and using cumulative-dose procedures (fourth row). The columns show data for individual pigeons. Abscissae: Doses of pentobarbital or amphetamine in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

Figure 2 shows the dose-response curves for morphine and methamphetamine for each pigeon in the methamphetamine–morphine group. As in Figure 1, both cumulative-dose and single-dose sessions are shown. At 1 mg/kg morphine, all 3 pigeons responded only on the saline-appropriate key. As the dose of morphine increased under the single-dose procedure, responding switched to the morphine-appropriate key at 3.0 mg/kg (Pigeons P404 and P411) or at 5.6 mg/kg morphine (Pigeon P412). The cumulative dose-response curve for morphine (second row) was similar to that obtained using single-dose testing procedures (first row). Similarly, at 0.3 mg/kg methamphetamine, all pigeons responded only on the saline key during single-dose test sessions (third row). At 1.0 mg/kg methamphetamine, Pigeons P404 and P411 responded on the methamphetamine key, but Pigeon P412 did not respond on the methamphetamine key until 1.8 mg/kg methamphetamine was given. Once again, the results obtained with methamphetamine during cumulative-dose sessions (shown in the fourth row) were similar to those obtained with single-dose sessions. Thus Figures 1 and 2 show that dose-response curves determined using single-dose and cumulative-dose testing procedures were strikingly similar.

Fig. 2. Patterns of responding on each key following increasing doses of morphine and methamphetamine in pigeons trained to discriminate among saline, morphine, methamphetamine, and a combination of morphine and methamphetamine (methamphetamine–morphine group).

The top two rows show morphine dose-response curves determined using single-dose procedures (first row) and using cumulative-dose procedures (second row). The bottom two rows show methamphetamine dose-response curves determined using single-dose procedures (third row) and using cumulative-dose procedures (fourth row). The columns show data for individual pigeons. Abscissae: Doses of morphine or methamphetamine in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

Figure 3 shows the effects of combining doses of pentobarbital and amphetamine for the amphetamine–pentobarbital group. Increasing doses of amphetamine (abscissas) in combination with increasing doses of pentobarbital (rows) were tested using single-dose procedures. The top row shows the effects of 1 mg/kg pentobarbital combined with different doses of amphetamine. When 1 mg/kg pentobarbital was combined with 0.3 mg/kg amphetamine, both of which had produced responding on the saline key when given alone, the pigeons also responded predominately on the saline key. When the dose of amphetamine was increased to 1.0 mg/kg in combination with 1 mg/kg pentobarbital, Pigeons P347 and P260 switched to the amphetamine key, but Pigeon P380 continued to respond on the saline key. Combinations of 1.8 or 3 mg/kg amphetamine with 1 mg/kg pentobarbital produced responding predominately on the amphetamine key in all pigeons. Thus the effects of amphetamine in pigeons also receiving 1 mg/kg pentobarbital were much like those when amphetamine was given alone (see Figure 1).

Fig. 3. Pattern of responding on each key following increasing doses of amphetamine in the presence of increasing doses of pentobarbital in pigeons in the amphetamine–pentobarbital group.

The columns show data from individual pigeons. Each row shows the dose-response effects of amphetamine in combination with a different dose of pentobarbital. Abscissae: Doses of amphetamine in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

The second row of Figure 3 shows the effects of 3 mg/kg pentobarbital combined with doses of amphetamine. After the combination of 3 mg/kg pentobarbital with 0.3 mg/kg amphetamine, Pigeon P347 responded on the saline key whereas Pigeons P260 and P380 responded on the pentobarbital key. Higher doses of amphetamine combined with 3 mg/kg pentobarbital produced responding largely on the combination key for Pigeons P260 and P380, but Pigeon P347 responded on the amphetamine-appropriate key at all of these dose combinations. When 5.6 and 10 mg/kg pentobarbital doses were combined with amphetamine (Figure 3, rows 3 and 4), pigeons responded on the pentobarbital key after the lower doses of amphetamine and on the combination-appropriate key after higher doses of amphetamine. Thus combinations of the lowest doses produced responding on the saline key. As the dose of pentobarbital increased in the presence of the lowest dose of amphetamine, responding shifted to the pentobarbital key. As the dose of amphetamine was increased in the presence of the lowest dose of pentobarbital, responding shifted to the amphetamine key. Combinations of higher doses of these two drugs produced responding predominantly, and often entirely, on the combination key.

Figure 4 shows the effects of combinations of methamphetamine and morphine for the methamphetamine–morphine group determined using single-dose testing procedures. At 0.3 or 1.0 mg/kg methamphetamine combined with 1.0 mg/kg morphine (Figure 4, row 1), responding was confined largely to the saline key. When higher methamphetamine doses were combined with 1 mg/kg morphine, all pigeons responded on the methamphetamine-appropriate key with Pigeon P412 shifting to the methamphetamine key at the 1.0 mg/kg dose of methamphetamine and the other 2 pigeons shifting to the methamphetamine key at 1.8 mg/kg of methamphetamine. When methamphetamine was combined with 3.0 mg/kg morphine (Figure 4, row 2), all the pigeons responded predominantly on the morphine key after the lowest dose (0.3 mg/kg) of methamphetamine was combined with morphine. Pigeons P404 and P411 switched to the combination key when the methamphetamine dose was increased to 1.0 mg/kg, but Pigeon P412 switched to the methamphetamine key at this dose before switching to the combination key at higher doses of methamphetamine plus 3 mg/kg morphine. All pigeons responded predominantly on the combination key when 3 mg/kg morphine was combined with the two highest doses of methamphetamine.

Fig. 4. Pattern of responding on each key following increasing doses of methamphetamine in the presence of increasing doses of morphine in pigeons in the methamphetamine–morphine group.

The columns show data from individual pigeons. Each row shows the dose-response effects of methamphetamine in combination with a different dose of morphine. Abscissae: Doses of methamphetamine in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

At 5.6 and 10 mg/kg morphine (Figure 4, rows 3 and 4), all pigeons responded primarily on the morphine key in the presence of 0.3 mg/kg methamphetamine. As the dose of methamphetamine increased in the presence of these doses of morphine, responding on the combination key predominated, except for the combination of 10 mg/kg morphine with 1.0 mg/kg methamphetamine in Pigeon P404, where responding on the combination key and on the morphine key were equally divided.

Thus combinations of low doses of morphine and methamphetamine produced responding on the saline key. As the dose of morphine increased in the presence of the lowest dose of methamphetamine, responding shifted to the morphine key. As the dose of methamphetamine was increased in the presence of the lowest dose of morphine, responding shifted to the methamphetamine key. Combinations of higher doses of these two drugs produced responding predominantly on the combination key. This pattern of interaction between methamphetamine and morphine was similar to the pattern of interaction between amphetamine and pentobarbital.

During training sessions after responding stabilized, mean overall rates of responding in responses per second (SEM) for the amphetamine–pentobarbital group were 2.03 (0.15) after saline, 2.17 (0.14) after pentobarbital, 1.19 (0.22) after amphetamine, and 1.65 (0.24) after the combination of pentobarbital and amphetamine. Results from a one-way analysis of variance (ANOVA) revealed significant differences in the rates of responding following the different training drugs during training sessions, F(3, 68)  =  5.28, p  =  .002. Rates of responding during the saline and pentobarbital training sessions were significantly higher than the rates of responding during the amphetamine training sessions (post hoc Tukey test). For the methamphetamine–morphine group, the overall rates of responding in responses per second (SEM) were 1.24 (0.13) after saline, 0.37 (0.36) after morphine, 0.81 (0.47) after methamphetamine, and 0.49 (0.29) after the combination of morphine and methamphetamine. Results from a one-way ANOVA revealed significant differences for rates of responding, F(3, 68)  =  17.09, p  =  .001, and a post hoc Tukey test showed rates of responding during saline training sessions to be significantly higher than rates during drug or drug-combination training sessions, and rates of responding during methamphetamine training sessions were significantly higher than rates during morphine and drug-combination training sessions.

Table 2 shows overall rates of responding during the drug combination tests. Both pentobarbital and amphetamine decreased rates of responding, especially at the higher dose levels. Similarly, both morphine and methamphetamine decreased rates of responding at the higher dose levels. The rate decreases produced by amphetamine were attenuated by pentobarbital, and the rate decreases produced by pentobarbital were attenuated by amphetamine. In contrast, the rate decreases produced by methamphetamine were not attenuated by morphine, and the rate decreases produced by morphine were not attenuated by methamphetamine. In fact, morphine added to the rate-decreasing effects of the highest dose of methamphetamine, and methamphetamine added to the rate-decreasing effects of the highest dose of morphine.

Table 2. Effects of individual drugs and drug combinations on overall rates of responding on all four response keys. Data are expressed as mean responses per second based on single observations in 3 pigeons.

Amphetamine doses (mg/kg)	Pentobarbital doses (mg/kg)
	0.0	1.0	3.0	5.6	10.0
0.0	2.03	1.11	1.46	1.30	0.28
0.3	1.54	1.38	1.03	0.84	0.84
1.0	0.57	0.99	1.10	1.14	0.65
1.8	0.61	1.16	0.85	1.22	0.71
3.0	0.04	0.60	0.56	1.04	0.29

Methamphetamine doses (mg/kg)	Morphine doses (mg/kg)
	0.0	1.0	3.0	5.6	10.0
0.0	1.24	1.03	0.44	0.26	0.37
0.3	1.60	0.31	0.58	0.34	0.21
1.0	0.58	0.55	0.31	0.35	0.15
1.8	0.81	1.46	0.18	0.14	0.11
3.0	0.42	0.16	0.14	0.16	0.05

Figure 5 compares single-dose session procedures (using some of the data from Figures 3 and 4) with cumulative-dose session procedures for the study of drug interactions. The top two rows of Figure 5 compare amphetamine–pentobarbital combinations under the two dosing procedures, and the bottom two rows compare methamphetamine and morphine combinations under the two procedures. The results are remarkably similar for the cumulative- and single-dose procedures, although there are some individual pigeon differences.

Fig. 5. A comparison of single-dose test procedures with cumulative-dose test procedures for the study of drug combinations.

The top two rows compare amphetamine and pentobarbital combinations under the two dosing procedures tested in the amphetamine–pentobarbital group. The bottom two rows compare methamphetamine and morphine combinations under the two procedures in the methamphetamine–morphine group. For both drug combinations, the first row shows results from single-dose procedures, and the second row shows results from cumulative-dose procedures. The columns show data for the individual pigeons. Abscissae: Dose of amphetamine (top two rows), or methamphetamine (bottom two rows), in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

Figure 6 shows the results from two determinations of cumulative dose-response curves for pseudoephedrine (top two panels) and nicotine (bottom two panels), two drugs to which these pigeons had not been exposed previously. Pseudoephedrine produced dose-dependent increases in responding on the amphetamine key in the amphetamine–pentobarbital group, reaching 100% amphetamine-appropriate responding at the 10 mg/kg dose, although Pigeon P380 failed to respond the first time tested after this dose of pseudoephedrine. The effects following doses of nicotine were similar in that low doses produced predominately saline-appropriate responding, and increasing doses produced dose-dependent increases in responding on the amphetamine-appropriate key. For nicotine, however, the highest dose tested, 3.0 mg/kg, resulted in partial substitution for the amphetamine stimulus with amphetamine-appropriate averaging 76%.

Fig. 6. Effects of pseudoephedrine and nicotine determined using cumulative-dose procedures in pigeons in the amphetamine–pentobarbital group.

The top two rows show pseudoephedrine dose-response curves determined on two separate occasions, and the bottom two rows show nicotine dose-response curves determined on two separate occasions. The columns show data for individual pigeons. NR indicates that the pigeon did not respond on any key following this dose. Abscissae: Doses of pseudoephedrine (top two rows) or nicotine (bottom two rows) in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

Figure 7 shows the effects of pseudoephedrine and nicotine during cumulative-dose sessions for pigeons in the methamphetamine–morphine group. During the first determination of the pseudoephedrine dose-response curve, after 1.0 mg/kg pseudoephedrine, responding occurred on the saline key for all 3 pigeons. As the dose of pseudoephedrine increased, responding on the methamphetamine key increased for Pigeons P404 and P411 before responding on the combination key predominated for Pigeon P404, but Pigeon P411 continued to respond on the methamphetamine key. Pigeon P412 responded only on the combination key after higher doses of pseudophedrine. No doses of pseudoephedrine resulted in much responding on the morphine-appropriate key. On the second determination of the effects of pseudoephedrine, all pigeons responded on the saline key at higher doses of the drug than had occurred during the first determination. Pigeon P404 responded on the saline key after all doses of pseudoephedrine, whereas the other 2 pigeons shifted to the methamphetamine key after the higher doses of pseudoephedrine. Thus pseudoephedrine had inconsistent effects in this group that depended on the pigeon and on the replication with responding on the combination key predominating in 2 of 3 pigeons after high doses of pseudoephedrine but which disappeared during replication of the dose-response curve.

Fig. 7. Effects of pseudoephedrine and nicotine determined using cumulative-dosing test procedures in pigeons in the methamphetamine–morphine group.

The top two rows show pseudoephedrine dose-response curves determined on two separate occasions, and the bottom two rows show nicotine dose-response curves determined on two separate occasions. The columns show data for individual pigeons. Abscissae: Doses of pseudoephedrine (top two rows) or nicotine (bottom two rows) in mg/kg. Ordinate: Percentage of responses on each key. Response-key designations are shown in the lower left corner of the figure.

Nicotine had unexpected effects in these pigeons. During the initial dose-response determination, all pigeons responded predominantly on the saline key after low doses of nicotine, but responding switched to the combination key as the dose of nicotine increased (Figure 7, row 3). Pigeon P404 showed similar effects during the replication with nicotine, and Pigeon P412 also switched to the combination key again after 3 mg/kg nicotine, although this pigeon responded predominantly on the methamphetamine key after the 1.8 mg/kg dose of nicotine. Pigeon P411, that had responded predominantly on the combination key after the two highest doses of nicotine, responded predominantly on the saline key after the 1.8 mg/kg dose of nicotine and on the methamphetamine key after 3 mg/kg nicotine, although some responding occurred on the combination key after both doses of nicotine. Thus there was considerably more responding on the combination key in the methamphetamine–morphine group after pseudoephedrine and especially after nicotine than had occurred in the amphetamine–pentobarbital group.

DISCUSSION

The present experiments extend the observations of McMillan & Li (2002) to new drug combinations, and they demonstrate the feasibility of using cumulative dosing procedures to study complex drug interactions in drug-discrimination research. When training drugs were given singly in the present study, responding was initiated on the saline key and shifted to the appropriate drug key as the dose of the training drug increased. When neither drug in a combination was given in a discriminable dose, responding was confined largely to the saline-appropriate key. When drug combinations were given with one drug in the combination at a discriminable dose (based on the dose-response curve for that drug when it was given alone) and the other drug in the combination at a nondiscriminable dose, responding occurred on the key associated with the discriminable dose. When drug combinations were given with both drugs in the combination given at a dose that was discriminable when the individual drugs were given alone, responding generally occurred on the combination-appropriate key. These findings held for 42 of the 48 drug combinations of pentobarbital and amphetamine and for 40.5 (with one drug combination, responding was equally divided between two keys) of the 48 drug combinations of morphine and methamphetamine tested. These simple rules account for 86% of the drug combinations studied in individual animals. Thus the present experiments show that the effects observed by McMillan and Li (2002) did not depend on an interaction between the particular drugs (morphine and pentobarbital) that were used in that study.

Beginning with Witkin, Carter, and Dykstra (1980), a number of investigators have studied the discrimination of combinations of pentobarbital and amphetamine; however, these investigations have usually been conducted using two-choice procedures. It is instructive to compare the findings in such studies with those of the present experiments. Perhaps the most interesting comparisons are between the present study and the two-choice procedures of Mariathasan, Garcha, and Stolerman (1991) and Mariathasan, Stolerman, and White (1999a), because similar training drugs were involved. Mariathasan and colleagues trained rats to discriminate a combination of 0.4 mg/kg of amphetamine and 10 mg/kg pentobarbital from vehicle. In a different group of rats, the same combination of amphetamine and pentobarbital was discriminated from its component drugs given alone. Rats trained to discriminate the combination from vehicle responded on the combination key following administration of either amphetamine or pentobarbital, but rats trained to discriminate the combination from the component drugs given alone did not respond on the combination-appropriate key when either of these drugs was given alone. Such findings suggested to these investigators that the discriminations were based on stimuli produced by the individual drugs rather than on a unique stimulus supplied by the drug combination, and that training the rats to discriminate the combination from the component drugs produced a higher level of specificity.

It could be argued, however, that animals trained to discriminate between the combination of the two drugs and saline learn to discriminate the presence of amphetamine and pentobarbital from the absence of these drugs. With only two keys available, when only one of the drugs in the combination is present, the animal responds on the key previously associated with that drug because no other option is available when only one drug is given. In our four-choice experiments, pigeons frequently responded on the combination-appropriate key even when the doses administered in the combination were doses that the pigeons had never been exposed to during training, as long as both drugs in the combination were of sufficient dose to be discriminated when given alone. If dose combinations that were never given during training were producing discriminative stimuli that were qualitatively different from those produced by the training combination, then it would be expected that the pigeons would respond on the saline-appropriate key. This did not happen. Thus our data support the conclusions of Mariathasan and colleagues.

When two drugs produce qualitatively different discriminative stimuli when the drugs are given alone, the stimulus effects of one drug might overshadow those of the other drug, especially if one of the drugs is given at a dose larger than the training dose (Mariathasan & Stolerman, 1993). This did not occur in the present study. If both drugs were given at doses that could be discriminated when given alone, the pigeons almost always responded on the combination-appropriate key when the drugs were given together, irrespective of the specific dose-ratio administered. In the current study, the pigeons responded on the combination key in 86% of the instances when two discriminable doses of the drugs studied were combined, suggesting that overshadowing is more likely to occur under two-choice procedures than under four-choice procedures where more response options are available.

In the present study, the interactions between amphetamine and pentobarbital and between methamphetamine and morphine were similar for drug discrimination; however, the interactions for rate of responding were different. Rate decreases produced by either amphetamine or pentobarbital were antagonized by the other drug, but rate decreases produced by methamphetamine or morphine were not antagonized by the other drug. In fact, drug combinations sometimes produced larger rate decreases than were observed when either drug was given alone. It appears that the determinants of drug interactions for drug discrimination are different from the determinants of drug interactions for rate of responding.

These pigeons required a year or more of training before stabilization occurred. Once responding stabilized, the study of large numbers of dose combinations required an additionally large number of test sessions. If a cumulative-dosing procedure could be employed to study the effects of drug combinations under the four-choice procedure, then such a procedure would greatly facilitate the rate of testing of new drug combinations. Therefore, we performed some tests with a dose of morphine or pentobarbital given once, followed by cumulative dosing with methamphetamine or amphetamine during a single session. The cumulative-dosing procedure generated data similar to the data obtained when single doses of each drug were combined during a test session (Figure 5). The use of cumulative dosing should greatly speed future experiments on drug combinations using the four-choice procedure.

We also studied the discriminative stimulus effects of pseudoephedrine in both groups of pigeons. Those in the amphetamine–pentobarbital group responded predominately on the amphetamine key after high doses of pseudoephedrine. In previous studies with rats and monkeys trained to discriminate amphetamine from saline using two-choice procedures, pseudoephedrine at least partially substituted for the training drug (Young & Glennon, 1998), or fully substituted (> 80% amphetamine-appropriate responding) in most of the subjects (Anderson et al., 2001; Tongjaroenbuangam et al., 1998). In rhesus monkeys trained to discriminate i.g. pentobarbital from saline, doses of pseudoephedrine did not produce any pentobarbital-appropriate responding. In the methamphetamine–morphine group, only 1 pigeon reliably responded on the methamphetamine-appropriate key after higher doses of pseudoephedrine. The other 2 pigeons responded primarily on the combination key after the 5.6 and 10.0 mg/kg doses of pseudoephedrine during the first cumulative dose-response-curve determination. During the replication, one of these pigeons responded only on the saline key after all doses, and the other switched to the methamphetamine key after 10 mg/kg pseudoephedrine. Whether these differential effects of pseudoephedrine reflect a difference between amphetamine and methamphetamine similar to that reported by Young and Glennon (1998, 2000), a difference between the drugs (pentobarbital or morphine) combined with amphetamine or methamphetamine when two drugs were combined, or differences resulting from other causes is not clear at this time. Nevertheless, pseudoephedrine substituted more consistently for amphetamine in the amphetamine–pentobarbital group than it did for methamphetamine in the methamphetamine–morphine group.

The effects of nicotine also were strikingly different in the two groups of pigeons. For those in the amphetamine–pentobarbital group, the effects of nicotine were much like those of pseudoephedrine. That is, low doses resulted in responding primarily on the saline-appropriate key, but as the dose of nicotine increased responding occurred largely on the amphetamine key. However, at the two highest doses there was responding on multiple keys, and approximately 20% of the responses were distributed on the saline-appropriate key, so that the maximum percentage of responding on the amphetamine key was 76%. Previously, we also observed partial substitution of nicotine for amphetamine in pigeons trained to discriminate between pentobarbital, amphetamine, and saline in a three-choice discrimination (Li & McMillan, 2003).

The effects of nicotine in pigeons in the methamphetamine–morphine group were different from the effects of nicotine in the pigeons in the amphetamine–pentobarbital group. These pigeons usually responded predominantly on the combination key rather than on the methamphetamine key after high doses of nicotine. The only exception was the replication study with Pigeon P411, and even this pigeon made a considerable number of responses on the combination key (30%, with the remainder being made on the methamphetamine-appropriate key) after 3.0 mg/kg nicotine. As was seen in the other group of pigeons, nicotine engendered responding on multiple keys, and overall, approximately 15% of responses were on the methamphetamine-appropriate key after the two highest doses of nicotine. Again, at this time it is not possible to determine if the differential effects of nicotine in the two groups of pigeons were caused by differences in the discriminative stimulus effects of amphetamine and methamphetamine, or differences caused by pentobarbital and morphine when they were given in combination with these two amphetamines.

One of the goals of this research program has been to use the four-key procedure to study drugs with discriminative-stimulus effects mediated by more than one mechanism. We reasoned that the stimuli produced by a drug with such complex actions should be analogous to a combination of drugs that have the discriminative properties of each of the components. For example, some opioid drugs have effects at both mu and kappa receptors. If pigeons were trained to discriminate among saline, a pure mu agonist, a pure kappa agonist, and a mixed agonist with both mu- and kappa-agonist effects (or a combination of the two pure agonists), then by varying the dose combinations of the mu and kappa agonists it should be possible to assess the relative contributions of mu and kappa activity to the discriminative stimulus produced by the mixed agonist.

In the present experiments using FR schedules, a wide range of combinations of amphetamine and pentobarbital or methamphetamine and morphine produced similar patterns of responding confined completely, or almost completely, to the combination key. Should similar effects be obtained, for example, with different combinations of mu and kappa agonists, it would be difficult to determine the relative contributions of these actions to the effects of a mixed mu and kappa agonist, because most dose combinations of the pure agonists would be expected to result in responding on the combination key. We have conducted a large number of experiments suggesting that FR schedules usually favor the development of quantal responding in drug-discrimination experiments, whereas graded responding is more likely to occur under interval schedules (see for example Massey, McMillan, & Wessinger, 1992; McMillan et al., 2001). Such experiments suggest that drug discriminations maintained under fixed-interval schedules might be more successful in quantifying the graded contributions of different mechanisms to the discriminative-stimulus effects of a drug with complex actions, although the point has yet to be demonstrated.

The present study, as well as that of McMillan and Li (2002), shows that pigeons can learn to discriminate among two drugs, a combination of these two drugs, and saline. The effects do not seem to depend on the particular drugs chosen because the procedure has been demonstrated with a number of different drugs. Furthermore, the four-choice procedure demonstrates that when a wide range of drug doses that are discriminable when given alone are combined, responding occurs on the combination key, even when the pigeon has never experienced these dose combinations during training. These effects were observed consistently under both single-dose and cumulative-dose procedures. Under the conditions of these experiments, overshadowing of one discriminable drug dose by a discriminable dose of another drug does not seem to occur as it does under other drug discrimination procedures with fewer response options. Finally, high doses of nicotine in the pigeons in the morphine–methamphetamine group resulted in a considerable amount of responding on the combination key, a result which could mean that nicotine was producing a stimulus more similar to that of the combination than that of the component drugs. It would have been difficult to uncover this effect using conventional drug-discrimination procedures.