Drug Use: Group Habits and Individual Learning

One conundrum about habitual drug use is that it generally takes place in social settings, yet many of the theories about drug use reduce it to individual learning. Bridging the gap between learning, habit, and reinforcement (psychological theories) and social dynamics and relations is a major problem facing a better science of addiction.

During the last two weeks of my graduate class in neuroanthropology, we’ve been reading about the neuroanthropology of addiction, as well as complementary material on extended cognition, habitual learning, and ritual. Today I want to highlight one way to link individual habits and cultural learning. This research comes from Luke J. Matthews and collaborators; Matthews did this research while on a post-doc at Harvard, and is currently the Director of Analytics at Activate Networks. So two papers:

Matthews, Luke J., Annika Paukner, & Stephen J. Suomi (2010). Can traditions emerge from the interaction of stimulus enhancement and reinforcement learning? An experimental model [full text]. American Anthropologist 112(2): 257-269.

Franz, Mathias & Luke J. Matthews (2011). Social enhancement can create adaptive, arbitrary and maladaptive cultural traditions [full text]. Proceedings of the Royal Society B: Biological Sciences 277(1698): 3363-3372.

The first paper covers the theoretical model, and experimental work with capuchin monkeys, that argues for an approach to cultural/group traditions that can emerge from simpler learning mechanisms than normally posited (no sophisticated imitation or conforming to a norm). Stimulus enhancement is a poor man’s version of imitation: “the stimulus-enhanced observer devotes more of its attention and activities towards an object or part of an object on which it saw the demonstrator act (2010:259).” This form of initial social influence can then shape subsequent individual learning, through patterns of reinforcement. They focus on looking at specific learning curves here, where higher pay-off curves might require more costs up front.

Their experimental work conforms to several of their predictions, thought they don’t show the crucial step – the establishment and transmission of traditions. Still, what caught my eye as a researcher on drugs was this specific statement:

Our model predicts that when individuals experience a more costly learning curve, they will focus more greatly on a variant that has already succeeded with some efficiency for them than on innovative behaviors because individuals will act on the only information they have and avoid the presumed higher opportunity cost of early trials for alternative variants (2010: 265)

This section caught my eye because one conundrum of addiction is why individuals often stay stuck on the habit of using, even when they can be aware that making a change in their life (i.e., innovating) might be a good idea. But the opportunity costs of change can be high, and drugs – whatever their costs – are a powerful and easy means of generating state change.

The second paper uses agent-based modelling to examine this approach to habits and traditions. I was particularly struck by their analysis of the emergence of maladaptive traditions. In evolutionary terms, this generally means with a lower fitness payoff than would be optimal. But the approach can work as an analogue for thinking about maladaptive behavior (a metaphor or analytic frame, not a direct parallel!). Maladaptive, in Matthews’ terms, simply means that the long-term payoffs of a particular approach are not as great as they could be.

Drugs, often fun at the start and costly if you get hooked on them, offer a similar style problem. Still, most people who try drugs don’t get hooked, and many group settings – like college, where a heavy group tradition of drinking exists – are remarkable for how people get involved in a specific behavior, and then change out of that tradition upon leaving the group.

So the long section on maladaptive behaviors, with emphases addded:

The mechanism that led to the emergence of traditions in groups of social enhancement learners appears to be the interaction between individual reinforcement learning and social enhancement as proposed by Matthews et al. (2010). By observing another individual, the observer becomes more likely to engage in the same behaviour as the observed individual. The increased practice of this behaviour then causes the observer to become more proficient in the behaviour, which increases the probability to perform the same behaviour in future. Such behaviours might be appropriately termed ‘habitual’ (Matthews et al. 2010).

These dynamics can lead to feedback effects between group members, because the frequency with which an individual performs a particular behaviour naturally affects the probability of observing an individual with this behaviour for other group members. As a consequence, in our simulations the more common behaviour in the group was amplified over time by feedback effects that are created by social enhancement learning (compare figure 4a,d). If the behaviour performed more frequently in groups of individual learners was the more profitable behaviour, then social enhancement led to the emergence of adaptive traditions (figures 4f and 5d).

However, if the more frequent behaviour in groups of individual learners was the less profitable behaviour, then social enhancement led to the emergence of maladaptive traditions (figures 4f and 5f). Individual learners tended to perform the more profitable behaviour when it was only slightly harder to learn than the less profitable alternative, but if the more profitable behaviour was much harder to learn then individual learners usually performed the less profitable, but easier, behaviour. Thus, the interaction of social enhancement with reinforcement-induced habit explains the emergence of adaptive and also maladaptive traditions in our simulations.

Previous research has focused on maladaptive behavioural traditions resulting from individuals relying on social information in the absence of available reinforcement trials, or from informational cascades in which the influence of social learning increases with an increase in demonstrations of a given behaviour (Bikhchandani et al. 1992; Giraldeau et al. 2002). By contrast, in our model maladaptive traditions emerge despite the integration of information that is acquired socially and individually. The key mechanism that allows the emergence of maladaptive traditions is that individual learning results in a learning curve (figure 1).

If a behaviour is hard to learn then several individual learning trails are required to obtain information about the maximum pay-off that can be obtained from performing this behaviour. If another behaviour is less profitable but much easier to learn, then individual learning can lead to individuals getting stuck on the less profitable behaviour. As a consequence, individual learners in our model only rarely performed the more profitable behaviour under these conditions (2011: 3368-3369).

Let us contrast that with one of the main theories today about how drug use transition from voluntary to habitual use, from the Everitt & Robbins (2005) paper, Neural systems of reinforcement for drug addiction: from actions to habits to compulsion (pdf).

Drug addiction is increasingly viewed as the endpoint of a series of transitions from initial drug use—when a drug is voluntarily taken because it has reinforcing, often hedonic, effects—through loss of control over this behavior, such that it becomes habitual and ultimately compulsive. Here we discuss evidence that these transitions depend on interactions between pavlovian and instrumental learning processes.

With the work by Luke Matthews, we can add in the importance of stimulus enhancement – due to both the setting and to what already established users are doing – that focuses new users on particular activities and drugs over other potential behaviors in that situation. Using individual learning processes, a group pattern can still be established – a tradition of drug use that characterizes a specific group or social setting. This approach provides a very different view than the individual disease-of-the-brain style of thinking that is currently so dominant.

It is also a view that augments my chapter on Neuroanthropology and Addiction that just came out in The Encultured Brain: An Introduction to Neuroanthropology.

My chapter focuses more on the pathology of addiction – on how wanting too much, craving, and compulsive involvement combine with habitual behavior and the chunking of “series of actions, experiences and cues derived from the patterns of everyday life of hardcore users (2012: 340).” Together, a deep wanting and a personally relevant habit comprise the destructive core of addiction, which works in conjunction with other aspects of drug use like the social and individual dynamics of control and self-regulation.

With the Matthews’ work, I can now add in a better analysis of the earlier stage of drug involvement. How do people become initially involved? How does that relevant habit first get established? That process combines the interaction of social and individual learning, as these papers demonstrate. And that process of social/individual learning exists in relation to other alternative behaviors and life options drug users might have. Or might not, given the opportunity costs they particularly might face, as well as the simple denial of other life options by powerful social actors.

I’m also intrigued by how this pattern relates to relapse, when using suddenly strikes the person as a potential option – viable and close – and staying sober, and all the rewards that might follow, seems quite distant. The opportunity costs are high, and stimulus enhancement – coupled with cues and stress – might tip the balance.

Just one final addition to make, from a paper by Bonnie and de Waal (2007), Copying with Rewards: Socially Influenced Foraging Decisions among Brown Capuchin Monkeys (pdf):

We investigated how rewards for both self and others influence the foraging choices of captive capuchin monkeys (Cebus apella)… Under all conditions, subjects more often explored the same box as the model than was expected by chance. Thus, without ever receiving a reward themselves or without seeing another receive rewards, subjects’ searches were directed at the box explored by another monkey. The tendency to match the model’s choice increased if the subject was rewarded.

As Matthews et al. (2010) put these results, Bonnie and De Waal (2007) showed that “the degree of social influence on behavior could be over three times greater when the observer was reinforced for the same action as they observed, which indicates the potential for interaction between social and reinforcement learning (259).”

And those reinforcements can be both social and individual as well – the pats on the back and roar of friends as a beer is drained, and the sense of relaxation and feeling good that the alcohol can also help stimulate. And from there, stimulus enhancement and social influence can shape reinforcement learning, and help keep a tradition of heavy drinking alive at both the group and the individual level.

This approach helps bridge the individual/biology and group/culture divide that separates a lot of work on drug use and abuse. The reason that matters is because for drug users, these scholarly divides simply don’t exist in their lives. The pattern of their use does.

Matthews Papers with Abstracts:

Matthews, Luke J., Annika Paukner, & Stephen J. Suomi (2010). Can traditions emerge from the interaction of stimulus enhancement and reinforcement learning? An experimental model [full text]. American Anthropologist 112(2): 257-269.

The study of social learning in captivity and behavioral traditions in the wild are two burgeoning areas of research, but few empirical studies have tested how learning mechanisms produce emergent patterns of tradition. Studies have examined how social learning mechanisms that are cognitively complex and possessed by few species, such as imitation, result in traditional patterns, yet traditional patterns are also exhibited by species that may not possess such mechanisms. We propose an explicit model of how stimulus enhancement and reinforcement learning could interact to produce traditions. We tested the model experimentally with tufted capuchin monkeys (Cebus apella), which exhibit traditions in the wild but have rarely demonstrated imitative abilities in captive experiments. Monkeys showed both stimulus enhancement learning and a habitual bias to perform whichever behavior first obtained them a reward. These results support our model that simple social learning mechanisms combined with reinforcement can result in traditional patterns of behavior.

Franz, Mathias & Luke J. Matthews (2011). Social enhancement can create adaptive, arbitrary and maladaptive cultural traditions [full text]. Proceedings of the Royal Society B: Biological Sciences 277(1698): 3363-3372.

Many animals are known to learn socially, i.e. they are able to acquire new behaviours by using information from other individuals. Researchers distinguish between a number of different social-learning mechanisms such as imitation and social enhancement. Social enhancement is a simple form of social learning that is among the most widespread in animals. However, unlike imitation, it is debated whether social enhancement can create cultural traditions. Based on a recent study on capuchin monkeys, we developed an agent-based model to test the hypotheses that (i) social enhancement can create and maintain stable traditions and (ii) social enhancement can create cultural conformity. Our results supported both hypotheses. A key factor that led to the creation of cultural conformity and traditions was the repeated interaction of individual reinforcement and social enhancement learning. This result emphasizes that the emergence of cultural conformity does not necessarily require cognitively complex mechanisms such as ‘copying the majority’ or group norms. In addition, we observed that social enhancement can create learning dynamics similar to a ‘copy when uncertain’ learning strategy. Results from additional analyses also point to situations that should favour the evolution of learning mechanisms more sophisticated than social enhancement.

Related Posts Plugin for WordPress, Blogger...
This entry was posted in Addiction, Learning. Bookmark the permalink.

7 Responses to Drug Use: Group Habits and Individual Learning

  1. Maia Szalavitz says:

    Interesting post!!! What’s also important here is that harm reduction can be a group norm: cultures can come to terms with a drug in ways that minimize addiction or other harms or maximize them. Unfortunately, the American mix of prohibition and the idea that the addict is “powerless” once he or she uses a drug means that we tend to wind up with drug cultures that enhance harm or don’t act optimally to reduce it, anyway.

  2. Rob says:

    So, at the risk of oversimplifying things, this post is detailing the sum of addiction theories as neurochemical based + social group reinforcement?
    It seems fairly obvious as there is little in the life of our internal chemistry, gene expression included, that is not affected or involved in a feedback loop with our external, often social, environment as well.

    • JS_MD says:

      You’re right Rob. At the risk of oversimplifying things: things affect other things. It seems obvious to the point that nothing is worth investigating. Ever. Hey…I wonder if Jimmy John’s is hiring?

  3. Luke says:

    Daniel, thanks for commenting on my work on your excellent blog. I lurk on it sometimes. I found your application of my work to drug use insightful, and something I had not thought of. One connection I did make at the time I conducted this research is that the model in the two articles might also explain, in a way similar to your drug use analysis, the persistence of different reproductive strategies. For example, we usually just vaguely wave our hands at ‘environmental differences’ to account for why some men end up being dads who invest in offspring and others chase more mates. What if part of these differences go along with my model because you have a substantial learning curve to climb to become an effective ‘dad’ or to be an effective ‘cad’? Now, once a man is effective in one strategy, he is in a different payoff situation than when he could perform neither strategy effectively, because to switch he is essentially giving up on fitness he could be achieving while he learns the alternative.

    If social learning is involved in starting people down one path or another, then this feedback mechanism of social learning and reinforcement learning might produce the kind of local subcultures that seem stuck in a certain reproductive pattern. This is one connection I had at least from seeing some presentations at the AAA back when I did this work. Those presentations essentially showed quite well that men and women display cross-cultural variation in proxies for reproductive strategies (number of partners, age of first intercourse, etc.) even after you control for all sorts of things like economics, education, geography, etc. Maybe we are not missing the relevant ecological variables? Maybe at least some of these dynamics are driving by interactions of social and reinforcement learning that can create basically opposite behavior patterns that yet are equally Darwinian-rational given the actors’ different fitness payoffs from their learned skills.

    • daniel.lende says:

      Luke, thanks for the comment. I like your points that learning strategies have costs, and that fitness-hill-climbing strategies should be individualized as well as populational. Optimality approaches generally don’t take into account the idea of “sunk costs,” yet I have often found that idea helps me think about phylogenetic and developmental dynamics in interesting ways. In your example, the pay-off situations change because of the sunk costs in one approach or the other.

      The second part of the comment is quite useful to me. I would likely rework the notion of “subculture,” which is a bit too bounded and homogeneous. Your approach helps get around that, because the stimulus enhancement and the reinforcement could be tied to settings, social actors, individual strategies, and the like, such that a relatively stable group/setting dynamic could be created, even if the individual members come in and out. In my work with drug users (from non-users to addicts to recovering folk), there is a really interesting suite of ecological behaviors that do drive the interactions of social and reinforcement learning. Your comment has helped me make an interesting connection to some of the previous evolutionary work I’ve done, and see how I might integrate that with this new approach that focuses more on cultural evolution, but still with a Darwinian and neuroanthropological perspective. Fascinating stuff. Actually has me excited to go back to some of that data and rethink it in some new ways.

  4. Sober Living says:

    This is a thought provoking article, one of the finest blogs on addiction habits. Just love it..

  5. Pingback: Learning Determines Choice OR What I Did During My Spring Vacation - Neuroanthropology

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>