A Closer Look at the Rat Park Experiment, Part 3

By Kenneth Anderson, MA

Part 1 reviewed some of the historical background which led up to the rat park studies. Part 2 reviewed the rat park studies themselves. Part 3 will look at where we have gone since.

image of rat to symbolize the rat park studiesFollow-ups to Alexander’s Rat Park Experiment

The July 5, 1985 issue of The Journal of the American Medical Association (JAMA) published a study by Michael A. Bozarth and Roy A. Wise on the toxicity of heroin and cocaine in rats. Subjects were 23 male Long Evans rats. All 23 rats were housed in solitary confinement in laboratory cages and fitted with catheters so that they could self-inject drugs by pressing a lever. The rats were given unlimited access to the drugs for 30 days. Eleven rats were in the heroin group; all eleven learned how to self-inject heroin. The amount of heroin injected by these rats stabilized after two weeks. The rats in the heroin group continued to groom themselves and maintained a stable body weight throughout the experiment. Four of the 11 rats (36%) had died by the time the 30-day experiment ended. The seven surviving rats were all healthy.

Only 10 of the 12 rats in the cocaine group learned how to self-inject cocaine. Nine of these 10 rats (90%) died in the course of the 30-day experiment. The cocaine rats ceased grooming themselves and lost large amounts of weight.

Of course, 1985 was the height of the Reagan drug war, and not long after the JAMA article appeared, the Partnership for a Drug-Free America released their notorious Cocaine Rat commercial, which states that 90% of rats which take cocaine die, and which seems to erroneously show the rat eating the cocaine.

1989 Study by Bozarth, Murray, and Wise

In 1989, Michael A. Bozarth, Aileen Murray, and Roy A. Wise published a study on the effect of housing (solitary confinement vs. group housing) on the self-injection of heroin or cocaine on rats. Subjects were male Long Evans rats which had been raised by the supplier in group cages; these rats were adults aged between 63 and 91 days old when they arrived at the laboratory, at which point they were either assigned to solitary confinement in individual cages or were placed in group cages of 10 rats each and were all fitted with catheters for self-injection. This group housing was substantially different than the group housing in the rat park experiments since it was all male and there is no reference to a sawdust floor or toys for the rats to play with.

There were two experiments in this study, one was a study of heroin self-injection and the other was a study of cocaine self-injection. Twenty-two isolated rats and 17 group-housed rats completed the heroin self-injection experiment. The experiment lasted for five weeks. For five days each week, the rats were removed from their cages and placed in a Skinner box for two hours and connected to a heroin supply which they could self-inject by pressing a lever. After the two hours were up, the rats were disconnected from the heroin supply and returned to their home cages. Every week, the rats had a two-day weekend when no testing was done. In other words, there were five blocks of five-day testing. Presumably, the rats did not become physically dependent on the heroin, since no withdrawal symptoms were reported. Isolated rats self-injected heroin more frequently than group-housed rats in all five testing blocks; however, this was statistically significant only in blocks two, three, and four.

Ten isolated rats and 14 group-housed rats completed the cocaine self-injection experiment. The procedure was the same as the heroin self-injection experiment. Although the isolated rats self-injected cocaine more frequently than the group-housed rats, this was not statistically significant in any of the five test blocks or in the test as a whole.

1996 Study by Bruce F. Petrie

In 1996, Bruce F. Petrie of the New Mexico Institute of Mining and Technology (New Mexico Tech) published a paper describing an attempted replication of the second rat park experiment. Subjects were 20 Wistar rats (10 male and 10 female). After weaning at 21 days of age, one group of rats was raised in isolated cages and one in rat park. Petrie does not specify the size of these two groups. The experiment began when the rats were 113 days old and lasted 28 days until the rats were 141 days old. The experiment was substantially similar to the second rat park experiment with a few small variations: instead of the eight phases of the original experiment, this experiment had seven phases, each lasting four days. This experiment also used a computerized measurement of fluid consumption. All the test solutions were 10% sucrose, and phase two used quinine instead of morphine. In phase three, the test solution contained 1.0 mg of morphine per milliliter of sugar water, in phase four this was 0.5 mg of morphine, in phase five, 0.25 mg of morphine and in phase six, 0.125 mg of morphine.

In phase three, the rat park rats consumed significantly more milligrams of morphine per kilogram of body weight than did the isolated rats. Differences were not significant in other phases of this experiment for this measure. This is the exact opposite of the results obtained in the second rat park experiment from 1979, where the isolated rats consumed more morphine. It is unclear whether this result was due to a genetic difference in the rats, or the different equipment used to measure consumption, or something else.

More contemporary research paints a far more complex picture than the simple impact of environment found in the original rat park study. For example, in 2001, Michael Thomas Bardo et al. published a study on self-injection of amphetamine by rats. Subjects were divided into three groups: the first group was raised in solitary confinement in individual cages, the second group was raised in social housing with eight to 10 same-sex rats in a single cage but with no toys or handling, the third group was raised in an enriched housing with eight to 10 same-sex rats in a single cage and with toys and daily handling by the experimenters. The rats were tested with two different doses of amphetamine: 0.03 mg/kg per injection and 0.1 mg/kg. When tested with the low-dose of amphetamine, rats from enriched housing injected significantly less than rats from isolated housing; there were no other significant differences with low-dose amphetamine. When tested with high-dose amphetamine, there were no significant differences between groups. This demonstrated that both dose and housing were important factors in how rats respond to drugs.

2007 Study by Rixt van der Veen, et al.

In 2007, Rixt van der Veen et al. published a study about the interaction of housing and genetics and how it affected the self-injection of cocaine by mice. Two genetically different mouse strains were used, C57 mice and DBA mice. Prior research had shown that these two strains react very differently to cocaine; the DBA mice react more strongly to cocaine but self-inject less. There was a total of 80 mice, 40 from each strain. The experimenters did not specify the gender of the mice. The mice arrived at the laboratory when they were eight weeks old and were all housed in isolated cages. Three weeks after they arrived, half of the mice of each strain were group-housed, four to a cage, for 19 days, after which, they were returned to isolated cages for eight days, at which point the cocaine self-injection experiment began. The group housing experience had a significant effect on the DBA mice; the DBA mice which had experienced group housing self-injected significantly more cocaine than the DBA mice which had lived in continuous isolation; the exact opposite of the results of the rat park experiment. The group housing experience had no effect on cocaine self-injection for the C57 mice.

2007-2009 Studies by Daniele Caprioli, et al.

Daniele Caprioli et al. published three studies about the effect of environmental change on self-injection of drugs by rats in 2007, 2008, and 2009. The first was a study of self-injection of cocaine with 83 male Sprague Dawley rats. The intent of the experiment was to see if there would be a difference in the amount of cocaine injected by rats who were injecting themselves in their home cage compared to rats who were injecting in an unfamiliar environment. All of the rats were raised in isolation; half were raised in Skinner boxes which served as their home cages and half were raised in regular cages. The Skinner boxes and the regular cages were for all intents and purposes identical. When the experiment began, the rats raised in Skinner boxes injected themselves in their home cages. The rats raised in standard cages were transferred to a Skinner box for the injection session, then transferred back to their home cages after the session ended. Injection sessions for both groups of rats lasted for three hours. Injection sessions took place once a day for 12 days. Rats who were in their home cages injected significantly less cocaine than rats who had been transferred from their home cage to a Skinner box, i.e., a novel environment.

In their second study (2008), Caprioli et al. studied heroin self-injection in rats using the same procedures they had used for cocaine and got exactly the opposite results. Rats which were injecting heroin in their home cages injected significantly more heroin than rats injecting in a novel environment.

In their third study (2009), Caprioli et al. trained rats to administer both cocaine and heroin separately, i.e., cocaine training and heroin training took place on alternate days. After training, the rats were given a choice of self-injecting cocaine or self-injecting heroin. The rats who were injecting in their home cages preferred heroin, whereas the rats who were injecting in a novel environment preferred cocaine. Caprioli et al. also interviewed humans who used both heroin and cocaine and found that most humans preferred to use heroin at home and cocaine outside their homes.

2008 Study by Marcello Solinas, et al.

A 2008 study by Marcello Solinas et al. found that an enriched environment could reverse aspects of cocaine addiction in C57 mice. Mice which have been habituated to cocaine or other drugs develop a preference for the environment where they received the drug; they tend to hang out more there as though they were waiting to get more of the drug. This is known as “conditioned place preference.” The standard apparatus for measuring conditioned place preference is a rectangular cage with grid flooring on one half, and hole flooring on the other half. If the mice received cocaine while on the grid flooring, they will prefer to hang out on the grid flooring, and vice versa. Another aspect of cocaine habituation is called “behavioral sensitization.” The first time that mice get a shot of cocaine they do not react very much. However, after they have been habituated to cocaine, they run around a lot after a shot. This is called “behavioral sensitization.”

Mice in the Solinas et al. experiment were housed four to a cage in standard housing until the experiment began. Then they were injected with cocaine once per day in the conditioned place preference cage for six days. The mice were next divided into two groups: one group was housed in standard housing, and the other group was housed in enriched housing which consisted of larger cages containing a running wheel and a small house and four to five toys that were changed once per week with new toys of different shapes and colors. Thirty days after the last shot of cocaine, the isolated mice continued to show conditioned place preference but the mice from the enriched environment did not. Likewise, when given a priming shot of cocaine, the isolated mice ran around a lot, but the mice from the enriched environment reacted the same as mice who had never had cocaine before. The mice from the enriched environment also showed no reinstatement of conditioned place preference after the priming shot of cocaine.

2019 Study by Walters and Kosten

In 2019, Hailey Walters and Therese A. Kosten published a review of studies on the development of addictions in animals which had undergone early life stress. There were 12 studies on alcohol, 13 studies on opiates, and 35 studies on cocaine; however, not all studies included self-administration, some were about conditioned place preference, etc.

Of the studies on alcohol, eight involved a choice between drinking water and drinking alcohol. Two of these were studies of maternal separation in Wistar rats: neither study showed a significant difference in alcohol consumption between these rats and control rats. Three were studies of maternal separation in Long Evans rats: two of these studies showed that the rats who had undergone maternal separation drank significantly more; one study showed no significant difference. A study of Swiss mice showed that maternal separation increased alcohol consumption relative to controls, as did a study of C57 mice. A study of Long Evans rats showed no significant effect of handling during early life.

Of the studies on opiates, four involved a choice between drinking morphine solution or water, and all were done with male rats. Three of these were studies of maternal separation: all three led to increased morphine consumption in adulthood compared to control rats. The fourth study showed that handling in early life led to increased morphine consumption in adulthood compared to controls.

Two of the stimulant studies involved a choice between water and a solution containing cocaine or amphetamine. Long Evans rats which had undergone maternal separation showed no significant difference from controls in consumption of cocaine or amphetamine. Wistar rats which had experienced early life handling initiated greater intake of cocaine earlier across days compared to controls.

Eleven of the stimulant studies involved self-injection. Ten of these studies involved maternal separation or neonatal isolation, i.e., separation from both mother and littermates. Some of these experiments resulted in increased acquisition or maintenance of the drug habit, and some resulted in the opposite–it depended on the strain of rat or mouse. One study involved early life handling of Sprague Dawley rats and resulted in enhanced acquisition of the drug habit.

2021 Study by Celine Nicolas, et al.

In 2021, Céline Nicolas et al. published a study of the effects of enriched environment on rats which had been self-injecting methamphetamine. Subjects were 20 adult male Sprague Dawley rats. On arrival at the lab, the rats were housed two to a cage in standard cages. After one week, self-injection catheters were implanted, and one week after that, the experiment began. Rats were allowed to self-inject methamphetamine 14 hours a day for five to six days per week, until 20 self-injection sessions had been completed. Injections were triggered by nose pokes. After the last self-injection session was completed, the rats were divided into two groups of 10 rats each. One group was housed three to a cage in standard cages. The other was housed three to a cage in much larger cages, each of which had a house, a running wheel, three floors connected by ramps or tunnels, and four toys that were changed once per week. After three weeks of abstinence, rats were placed in the same chambers where they previously self-injected methamphetamine. The rats from the enriched environment showed significantly fewer methamphetamine-seeking nose pokes than did the rats from standard housing.

The experimenters also measured the amount of glucocorticoid receptor expression in the rats after the abstinence period: glucocorticoid receptors are receptors for stress hormones. Enriched housing during the abstinence period had normalized glucocorticoid receptor expression in the hippocampus of the rats. Standard housing had not.

2021 Paper by Lidia Belles, et al.

A 2021 paper by Lidia Bellés et al. reported a study of the effects of enriched environment on addiction-prone rats (Roman high-avoidance rats) and addiction-resistant rats (Roman low-avoidance rats). This study looked at cocaine self-injection, impulsivity, novelty preference, amphetamine-induced dopamine release, and the availability of dopamine D2 and D3 receptors. All rats studied were male. After weaning, 14 addiction-prone rats were raised in an enriched environment, 14 addiction-prone rats were raised in an impoverished environment, 14 addiction-resistant rats were raised in an enriched environment, and 14 addiction-resistant rats were raised in an impoverished environment, making a total of 56 rats. All rats took about an equal amount of time to learn to self-inject cocaine. But surprisingly, the addiction-prone rats from the enriched environment showed a faster rate of cocaine self-injection and greater cocaine intake than the addiction-prone rats from the impoverished environment. The addiction-prone rats from the enriched environment also showed greater impulsivity than the addiction-prone rats from the impoverished environment, and a greater density of dopamine D2 and D3 receptors on the dorsal striatum; however, they unexpectedly showed lower novelty preference. No environment-related difference in rate of cocaine self-injection or cocaine intake was found in the addiction-resistant rats. Since novelty seeking and impulsivity are associated with addiction proneness in humans, these results were surprising.

2022 Study by Paola Maccioni, et al.

In 2022, Paola Maccioni et al. published a study on alcohol self-administration by 36 male Sardinian alcohol-preferring rats; these are rats which have been especially bred to consume large amounts of alcohol. When the rats were 21 days old they were each assigned to one of three living conditions: 12 rats were housed in solitary confinement in an impoverished environment comprised of a shoebox-like cage, 12 rats were housed in a standard environment with three rats to a cage, and 12 rats were housed in an enriched environment with six rats to a cage in a large cage with two stories, running wheels, a maze, etc. Until the rats were 50 days old, they were given just food and water, but at 50 days old they were given a choice between water and a 10% alcohol solution in their home cages. When the rats were 60 days old, they were placed in a skinner box for a part of each day and taught to self-administer water by pressing one lever or alcohol solution by pressing another lever. The skinner box had two drinking cups; pressing the water lever released one drop (0.1 ml) of water into the water cup and pressing the alcohol lever released one drop (0.1 ml) of alcohol solution into the alcohol cup. During the training period, rats from the enriched environment took significantly longer to learn to self-administer alcohol than the rats from the impoverished environment and drank significantly less, indicating that they were less motivated to consume alcohol. Rats from the standard environment scored in between the other two groups. During the testing period, the rats from the enriched environment once again consumed significantly less alcohol than the rats from the impoverished environment. During the final testing phase, the experimenters progressively increased the number of lever presses required to obtain alcohol to see which rats would give up first. Not surprisingly, the rats from the enriched environment had the least interest in alcohol and gave up sooner than the other rats.

There are many more studies on the effect of environment on addiction in animal models which I have not covered here; I have tried to give a representative sample. One criticism of the animal studies done to date is that the overwhelming majority of these studies use same-sex rats or mice, and this is unnatural since in nature rodents are not celibate but have sex with partners of the opposite sex.

So, what does Rat Park mean for us?

The rat park story should serve as something of a cautionary tale. As H. L. Mencken said, “Explanations exist; they have existed for all time; there is always a well-known solution to every human problem–neat, plausible, and wrong.” Rat park fired the imaginations of popular writers on addiction with the idea that social reform could end addictions among humans. There is good evidence that there is a strong relation between social conditions and addiction in humans, for example, the 2021 National Survey on Drug Use and Health (NSDUH) reports that rates of drug use disorder are about twice as high (12.7% vs. 6.8%) among the very poor (less than 100% of the poverty level) as they are among the non-poor (greater than 200% of the poverty level). Interestingly, rates of alcohol use disorder among these two socioeconomic groups are about the same (10.5% vs. 10.8%).

However, Bruce K. Alexander’s 1978 rat park study just happened to fortuitously hit on the right combination of genetics, drug, and environment to produce the results that it did. More contemporary research tells us that vulnerability to addiction in animals is the result of a very complex interaction of multiple factors, including early environment, current environment, maternal nurture, genetics, gender, drug type, dosage, and route of ingestion. Nevertheless, the rat park studies were crucial for advancing and inspiring research into the complexities of the effects of environment on addiction.