Drug discrimination (DD) test is a type of behavioral testing in which animals are trained to associate a particular drug with a particular stimulus. In a classic operant drug discrimination paradigm, animals learn to discriminate against a novel test drug from a reference agent. DD studies are performed to analyze in vivo pharmacological effects of a novel drug of abuse and distinguish them from the effects produced by another drug or in the absence of a drug. During the procedure, the subjects are trained to associate drug administration/dosage with a response that ultimately leads to reinforcement. In this way, the animal seeks to identify the drug dosage and effects produced by the drug. This drug is referred to as a “training drug.”

In a typical DD paradigm, the subjects are trained in a way that when they are administered a particular dose of a drug of choice, they press the lever for reinforcement. The drug can be administered alone or in combination with the training drug to assess which lever will the animal consequently press. DD test is vital for analyzing the abuse potential of drugs because it explains the interaction of test compounds with CNS. After all, if it fails to cross the blood-brain barrier, it cannot be used as a discriminative stimulus. In addition, the DD assay serves as a consistent model of the subjective properties of the discriminative stimulus (drug of choice). Discriminative drug procedures can be employed to investigate features like genetic manipulations, pharmacokinetics, behavioral history, neurobiological, and interoceptive effects of novel test drugs (Solinas et al., 2006).

Drug discrimination assay comprises four basic components i.e., subject, discriminative stimulus (drug dosage), an appropriate response, and presentation of reward (Young, 2009).

In a typical DD procedure, animals such as rodents are subjected to short training sessions of 15-30 minutes in which they learn to associate lever pressing with reinforcement in the form of the presentation of food pellets. The subjects are daily injected with a training drug or its vehicle, placed in the operant conditioning chamber, and allowed to select between two levers. Responses on one lever are associated with reinforcement during drug sessions, whereas responses on the other are associated with reinforcement during vehicle sessions. Once the subject chooses the correct lever, test sessions can be conducted (Young, 2009).

Testing can be done 1-2 times a week whereas baseline sessions are scheduled between the test sessions. Test sessions help the researchers analyze (1) whether the test drug mimics the action of the training drug, and (2) whether the treatment alters the effects of the training drug. The animal’s behavior helps measure the number of responses on drug-appropriate lever and response rate.

Dosage effect curves can be plotted and response rates can be measured by varying the dose of the training drug during the test sessions.

Fixed Ratio (FR)

The fixed ratio represents the number of responses required to obtain reinforcement such as food pellets. The subject is presented with reinforcement only after completing a fixed number of responses and cannot be rewarded before that. For instance, in a FR 10 schedule, you can reinforce the subject on every 10th response.

The experimenter can increase the sensitivity of the DD assay by increasing the number of responses per reinforcement (e.g., food pellets). The FR value is random, but most DD paradigms employ FR10 schedules. It minimizes the chance of involuntary presses disrupting the measurement and at the same time can be quickly produced by rodents.  The subject must make 10 consecutive responses on the correct lever without responding to the inactive lever to fulfill the FR requirement. To accomplish this, the scientists can reset the acquired responses on the active lever keeping in view the number of presses on the inactive lever.

Variable Interval (VI)

In a variable interval (VI) schedule, the researchers can vary the length of time between two consecutive reinforcements. For example, if you deliver the first reinforcement 15 seconds after the response, increase the time by 2 seconds for the next reinforcement. Once the subject learns to press the right-side lever and left-side lever, initiate DD training.

Some researchers also use FR and VI in combination but FR is the most frequently utilized schedule in DD experiments.

Number of Trials per Session

A trial is defined as a session’s sub-part that is dismissed as soon as the reinforcement is delivered. The number of trials per session is random and selected based on experimental requirements.  The experimenters usually use 20 trials per session. In addition, the number of trials is kept constant from initial shaping to the end of the DD paradigm. However, it can be varied depending on the nature of the experiment.

Time Out

Time out is defined as the duration after delivering reinforcement in which the presses on the active lever do not affect the consequences. During TOs, the subject receives and consumes food pellets delivered as reinforcement. TO values are shorter during shaping lasting for about 5s whereas longer TOs of about 30-the 60s are recommended during the final schedule.

Length of Sessions

The maximum length of DD sessions depends on the pharmacokinetic properties of training drugs, and the researchers should consider underlying aspects while determining the length of sessions.

•       Sessions are terminated on reaching the maximum number of trials.
•       Maximum session duration should be the same for both training and test sessions.
•       Sessions should be long enough to allow the maximum number of trials to be completed under baseline conditions.
•       Sessions should not be too short that subjects are unable to yield a sufficient number of responses.
•       Extremely long sessions are insensitive to short-acting drugs.

Range of doses

The range of doses defines characteristics like behavioral effects and selectivity of the test drug and can be determined by literature review. The dosage range should extend for at least a logarithmic unit like 2-20, 4-40, and so on. Once the range is decided, the doses should be administered non-systematically. If the dosage range decreases the responding rate in subjects or has toxic effects, it should be avoided. Similarly, very high doses with increased toxicity should not be directly tested on animals.

Despite these toxic effects, the researchers might need to test high doses based on experimental requirements. For instance, to find out the receptors associated with the production of discriminative effects of drugs, high doses are required. In this case, it is advised to employ the range of doses within which the test drug is selective. However, to check the similarity between the discriminative effects of training and test drug, one must use high doses for testing that significantly decrease drug response.

Pretreatment Times

Pretreatment time should be determined based on the fact that discriminative effects of the test drug retain throughout the session. It facilitates the onset of drug effects at the beginning of the sessions and allows it to persist by the session’s end.

Type of Discriminative Responding

Discriminative responding can be of two types: graded or quantal. Quantal responding is all or none i.e., whether the discriminative effect is similar to the drug or not. In the case of quantal responding, lever selection is the main concern. The subject is trained to press one lever and it keeps pressing the same lever throughout the session. Usually, FR schedules are used in quantal responding. On the contrary, in the case of graded responding, the discriminative responding could lie anywhere between 0 and 100%. Graded-like effects are more likely observed when VI or VI-FR schedules are followed during the DD paradigm (Solinas et al., 2006).

Standard two-lever operant conditioning chambers are usually used for drug discrimination paradigms. The apparatus comprises two retractable levers, a device for delivering reinforcement, and two light stimuli. The lever can be placed on either side of the chamber and one light stimulus is placed above each lever.

One wall of the chamber can be opened for introducing the animal into the paradigm, whereas the other wall is equipped with two levers and a device placed between two levers at an equal distance from both of them. It is used to deliver reward/reinforcement to the animals placed in the chamber. The main operant conditioning paradigm is housed within outer chambers for attenuating light or sound. Each chamber is illuminated by a 28V overhead house light source. The equipment facilitates lever retraction, and reinforcement delivery, and records the number of reinforcements.

The researchers can employ underlying general protocols for performing drug discrimination tests (Solinas et al., 2006).

Operant Training Phase

1.     Initially train the animals for lever pressing using “magazine training.” Remove the levers from the operant conditioning chamber, place the rats in it, and allow them to get habituated to the equipment. Allow the subject to move randomly and deliver food pellet without associating it with the animal’s behavior. In this way, the animal will learn to associate the sound of the food dispenser with food presentation. After the training session, reinstall the levers and reinforce the response on either lever with food.

Alternatively, place the animals in the pellet with the levers installed and associate the food pellet as reinforcement with the response on either lever. Once the subject starts responding, you can gradually increase FR and TO.

2.     During the first few sessions, reinforce the animal as much as possible to encourage lever pressing. But make sure that the animal does not develop a habit of pressing only one lever as some experiments require the subject to press one lever and some require them to press the other one. To achieve this, remove one lever from the chamber and train the subject to press one. Once the subject learns this behavior, remove this lever, install the other, and reinforce responding on this one. Following this, install both levers simultaneously.

Alternatively, install both levers simultaneously and reinforce responding on each lever on alternate days. 

3.     Keep changing FR and TO values based on the subject’s performance. Start with a low FR value and keep increasing it in the subsequent sessions. Increase FR values within the sessions. For instance, if a subject’s response rate is reliable under FR2 for two to three consecutive sessions, increase the FR value to 3, and so on. Make sure that the subject consumes all available rewards within one session.

Increase TOs within the sessions as they do not require individual attention.

Note: If the subject’s response rate does not increase under increased FR value, reduce FR to the value at the start of the session.

4.     Monitor each subject’s behavior and set the schedule parameters independently for each subject based on their performance. Try to conduct the DD experiment over a small group of animals instead of larger ones as it will be difficult to pay attention to individual subjects in large groups.  

Drug Discrimination Training Phase

5.     Once the subjects acquire operant behavior, start injecting the drug. You can use drug versus vehicle training in the beginning and once the operant behavior stabilizes and the animals start responding reliably and later, increase TO and FR to their final value. Keep the reinforcement rate on the correct lever high to help the animals to learn discriminative behavior.
6.     Administer the drug and vehicle with the same frequency to make sure that the animals do not develop a preference for one lever but do not make the order of vehicle and drug session too obvious. You can conduct 2 days of drug sessions followed by two days of vehicle sessions for this purpose.
7.     Replace the animals that fail to acquire discrimination within 3-5 months of training duration.

Testing and Maintaining DD

8.     After the acquisition of discrimination, you can use two criteria to assess it.
9.         Percent correct responses during the session
10.         Number of incorrect lever presses during the first trial

Greater than or equal to 80% of correct responses allows a good assessment of drug effects. However, if you use the FR10 schedule, preferably consider the “number of incorrect passes during the first trial” to assess DD. If there are fewer than five presses on the incorrect lever, the DD training criteria is being satisfied.

9.     After fulfilling the training criteria, administer multiple doses less than the training drug and plot a curve between dosage and drug effect. These dose-effect curves are called “generalization gradients.” These gradients are S-shaped graphs and vary depending on drug characteristics and dosage. Plot generalization curves at the beginning and the end of the DD experiment, or every 5-6 months.

Note: You can lower the dose to increase vehicle selection.

10. Test different doses of drugs in a “semi-random manner.” and constantly evaluate the effects of training drug dosage and its vehicle. Also, calculate the range of the training dose logarithmically.
11. Conduct test sessions twice a week, say Tuesdays and Fridays, and baseline sessions for the rest of the weekdays. DD experiments can be continued for five days per week.

Various applications of drug discrimination tests have been developed by now. For instance, a blockade test performed using DD assay is used to analyze how treatment alters the effects produced by the training drug. Similarly, the DD test can be used to study how neurotransmitters interact with the effects produced by the test drug. It can also be employed to study the pharmacokinetic effects of discriminative drugs. However, one application of the drug discrimination paradigm is discussed below.

Generalization Test

Walentiny et al. (2019) conducted experiments to study “oxycodone-like discriminative effects of fentanyl-related emerging drug abuse in mice.” The researchers took C57BL/6 mice (adult males) and trained them to discriminate 1.3mg/kg oxycodone from vehicle drugs. They employed standard two-lever operant conditioning chambers for this purpose. The operant conditioning paradigms were housed in light- and sound-attenuated boxes. The front panel of the apparatus contained two retractable levers with a food dispenser centered between them whereas the rear panel was equipped with a light and a fan.  Initially, the experimenters conducted a 15 minutes operant learning session in which mice were trained for pressing one lever using the FR1 schedule. Consequently, the animals were reinforced with a 20mg food pellet. The FR value was gradually increased to 10 after which they trained the mice for pressing the opposite lever, in the same manner, using the FR10 reinforcement schedule. After the mice had established the operant behavior, the scientists administered them with a 1.3mg/kg training dose of oxycodone or saline (vehicle) and placed them in the chamber after 15 minutes for a 15 minutes “errorless” training session. During this session, the subjects pressed only the drug-appropriate lever and 16 such sessions were conducted (8 with vehicle and 8 with drug).

Following this, the experimenters conducted DD training in which they injected the mice either with saline or oxycodone training dose in a semi-random manner (DDVVDD), thereby commencing 5 training doses and 5 vehicle training sessions over 2 weeks. During operant learning, both levers were extended and only responses on drug-appropriate levers were rewarded. The rest of the operant training and DD procedure was similar to the one described above.

Then, mice were pretreated with 1mg/kg naltrexone for 25 minutes, oxycodone for 15 minutes, and fentanyl and related compounds for 10 minutes) and then subjected to generalization tests. During generalization tests, the mice were injected with a dose of a given compound (ocfentanil, valerylfentanyl, crotonylfentanyl, 3-furanyl fentanyl) and vehicle. Then, the scientists placed the subjects in the chamber for a 15-minute test session. Then they recorded dose-response curves for each compound and calculated the “percent oxycodone lever responding.” The researchers concluded that “fentanyl and each of its analogs were completely generalized to 1.3 mg/kg oxycodone discriminative stimulus and naltrexone pre-treatment significantly decreased oxycodone-like responding for each compound.”

Drug discrimination paradigms have manifold advantages. The first benefit of DD assays is that they are easy to use. Secondly, such procedures are highly sensitive with distinctive molecular specificity. Drugs with common action sites produce similar discriminative effects. For example, if a subject is trained to discriminate CB1 receptors, only CB1 receptor agonists will produce full generalization whereas only CB1 antagonists will block the discriminative effects of THC. In addition, DD assays provide complete information regarding in vivo mechanisms controlling the psychotropic effects of drugs (Solinas et al., 2006).

A significant advantage of the drug discrimination paradigm is that the researcher can test the animals repeatedly for several years without apparent tolerance against discriminative drugs or loss of pharmacological sensitivity. Another advantage of DD techniques is that they do not depend on the subject’s responding rates. Drugs that interrupt responding rates cannot be used for discrimination assays. Last but not the least, “high predictive validity” is a chief benefit of drug discrimination paradigms.

Despite being quite useful, the technique has a few disadvantages as well. For instance, a potential disadvantage is that the sessions are quite long lasting up to several months. Moreover, the subjective effects of the drug and the subject’s discriminative abilities might change over time. However, the advantages of the DD paradigm have surpassed its disadvantages.

• Always test different drug doses in a “semi-random” fashion.
• During conducting blockade tests, a high dose of training drug can decrease the responding rate.
• Avoid abrupt routine changes as they can disrupt discriminative behavior such as responding rates. Gradually change the routine if the subjects have to be acclimated to the new routine.
• In case of relatively complex and repetitive tasks, prepare test tables including details like a drug, dose of the drug, time, route of administration, subject’s weight, and so on for your ease.
• Follow animal welfare and ethical guidelines while performing drug discrimination tests.

• Drug discrimination (DD) test is a type of behavioral testing in which animals learn to discriminate against a novel test drug from a reference agent.
• In a typical DD paradigm, the subjects are trained in a way that when they are administered a particular dose of a drug of choice, they press the lever for reinforcement. The drug can be administered alone or in combination with the training drug to assess which lever it will press.
• Drug discrimination assay comprises four basic components i.e., subject, discriminative stimulus (drug dosage), an appropriate response, and presentation of reward.
• Standard two-lever operant conditioning chambers are usually used for drug discrimination paradigms.
• Two types of reinforcement schedules are followed in the DD technique: fixed ratio (FR) and variable interval (VI).
• The fixed ratio represents the number of responses required to obtain reinforcement such as food pellets.
• In a variable interval (VI) schedule, the researchers can vary the length of time between two consecutive reinforcements.
• The main parameters affecting drug discrimination assays include the number of trials per session, time out (TO), pre-treatment times, length of sessions, and range of doses.
• DD paradigms are easy-to-use, highly specific, and do not depend on the subject’s responding rate.
• Various applications of DD assay include generalization tests, blockade tests, and studying the in vivo pharmacokinetic characteristics of discriminative drugs.

1. Solinas, M., Panlilio, L. V., Justinova, Z., Yasar, S., & Goldberg, S. R. (2006). Using drug-discrimination techniques to study the abuse-related effects of psychoactive drugs in rats. Nature protocols, 1(3), 1194-1206.
2. Walentiny, D. M., Moisa, L. T., & Beardsley, P. M. (2019). Oxycodone-like discriminative stimulus effects of fentanyl-related emerging drugs of abuse in mice. Neuropharmacology, 150, 210-216.
3. Young, R. (2009). Drug discrimination. Methods of behavior analysis in neuroscience, 2.