Puts me in mind of a study a few years back where they had socially housed monkeys self-administer cocaine [1]. Like all primates, each group of monkeys ended up with a dominance hierarchy. The most dominant had a relatively nice life compared with the subordinates: more grooming from others, more play, more attention in general.
It turned out that the subordinate monkeys were much more susceptible to the temptations of cocaine. In contrast, the dominant ones were able to resist its allure a lot more effectively.
The researchers also looked at the neurobiology of the monkeys. Before being socially housed, they were kept in individual cages for a couple of years. They looked at the main pleasure centre of the brain and measured the level of dopamine receptors (of the D2 type) present there. These receptors are directly stimulated by naturally occurring rewards like social activity, food, sex, and so on - but also stimulant drugs like cocaine.
All the monkeys had pretty much the same D2 receptor level when they were cooped up in single cages, with little variation. But when they were socially housed, the D2 levels of the dominant monkeys rose significantly higher; this too was associated with their lessened vulnerability to cocaine.
So it seemed that social environment can have quite the effect on the neurobiology of one's reward pathway, and potential for drug abuse (as the Rat Park experiment suggested).
On reward pathways - in human stimulant addicts you also see lower D2 receptor levels than in non-addicts [2], although this is lacking a comparative reading from before they were addicted, so we don't know if that's near the state they were in when they started abusing the drug. Having said that, in non-addicts there is a natural variation in D2 receptor levels, and those with less D2 enjoy stimulants a lot more [3], just like the subordinate, lower D2 receptor monkeys.
Unlike Rat Park, however, all the above is based on stimulants, which directly affect the aforementioned receptors. Whereas morphine takes a more indirect route, with different receptor types, so it's not quite comparable. Still, interesting to think about how much we may be slaves to our neurobiology, and our social surroundings.
[1] doi:10.1038/nn798
[2] doi:10.1177/026988119901300406
[3] doi:10.1176/ajp.156.9.1440