Addiction can be studied empirically from a dizzying number of angles. Scientists study how it is realized in the brain, its costs to society, what makes a person susceptible to it, how to treat it, and so on. Philosophers, of course, are also interested in addiction. In large part, this may be because addiction so clearly straddles what Wilfred Sellars called the scientific and manifest images of the world: the image of the world as presented to us by our best sciences, on the one hand, and the rich, normatively laden framework in which persons and their attitudes and reasons are central, on the other. Science provides a highly detailed picture of addiction at many different levels, but ultimately, we may need to somehow cash it all out in personal and interpersonal terms.
The traditional framing of the central question about addiction is the opposition between disease and moral failing. Among philosophers, this framing is now widely acknowledged to be mistaken, not least because such an opposition is pretty clearly both artificial and non-exhaustive. Even setting that aside, if what we want to know is how we should modulate our judgments, feelings, and actions concerning those who have an addiction, we should just ask those questions directly rather than trying to divine answers to them through a matrix of controversial assumptions about the nature of disease.
So, this is what we should do. Addiction clearly has effects at the level of persons and is therefore relevant for our assessment of persons. Can we draw on the plethora of probative science to determine how it has those effects and how it should affect our assessments?
There are many sciences of addiction that are relevant for this task: animal studies seem to show that rats in stimulating and enriched environments take drugs at lower rates than caged rats; epidemiology teaches that most diagnosable cases of substance abuse and dependence resolve before middle age; and behavioral science shows robustly that people with addictions respond to incentives. Results like these constrain the space of available views at the moral-psychological level: whatever motivational states are involved in addiction, and however they interact with the rest of a person’s mental economy, they don’t appear to operate by the force of literal compulsion.
More than any other science, however, the neuroscience of addiction seems to be especially relevant. We know that addictive drugs are addictive in large part due to their pharmacology and the changes that long-term exposure to them can cause in the brain. We also know that addictive drugs all increase dopaminergic activity in two of the brain’s main dopaminergic pathways, the mesolimbic and mesocortical pathways (together often referred to as the mesocorticolimbic dopamine circuitry), which are widely thought by researchers to be key to addiction. Crucially, it is also well-established that the mesocortocolimibic system not only responds to rewards, but to cues that reliably predict rewards. Broadly speaking, there are two leading neuroscientific theories of addiction that build on the key insight that one function of dopamine transmission is to encode information about relations between cues and rewards.
The prediction error theory says that addictive drugs interfere with learning by causing drug-taking to acquire, effectively, ever-increasing value in an animal’s representational system. According to this theory, the dopamine signal induced by addictive drugs is a learning signal.
The incentive sensitization theory says addictive drugs boost the strength of already established motivational associations. On this view, the dopamine signal induced by addictive drugs is a salience-boosting signal; addiction is a form of disordered motivation, not a form of disordered learning.
The relation between the two theories is contentious. They are typically considered competitive, but the two theories may be complementary. (See my paper here for an argument suggestive of this conclusion.) But, even if that is not true, the incentive sensitization theory is a leading contender worthy of consideration in its own right. For these reasons, I will focus on the incentive sensitization theory.
Some of the key evidence for the incentive sensitization theory comes from a paradigm known as Pavlovian instrumental transfer. For instance, if rats are trained to associate both a sound and an activity such as lever-pressing with a sugar reward, they will learn to associate the sound and the lever-pressing with one another, so hearing the sound will tend to trigger the lever-pressing. Using this paradigm, it is possible to test whether a given manipulation modulates the strength with which the rats “want” the reward by measuring the effect the manipulation has on how much the rats press the lever in the presence of the sound. It turns out that if you give rats amphetamine (which increases dopamine neurotransmission), they will press the lever up to three times as much upon hearing the sound, apparently wanting the reward more strongly. Strikingly, nearly identical results were found when the rats were not given amphetamine during the test but instead were given an escalating regime of amphetamines that ended weeks before the trial. In neither of these sets of experiments did the rats show signs that their liking of the sugar reward was enhanced. (Researchers measure liking by using a “taste reactivity measurement of hedonic impact,” i.e., they look at rats’ facial expressions and mouth movements.)
Researchers in this program take results such as these to support three claims: (i) the primary effect of the drug-induced dopamine signal is to drive up the motivational strength of cue-reward relationships; (ii) this effect is not active only when the drug is present in the animal’s system, but is stable for some time afterward; and (iii) the wanting that an animal is disposed to experience by this signal can come apart from how much the reward is liked.
If we now consider what happens when an addictive drug is itself the reward, we can begin to see how drug-taking can be self-reinforcing. Taking the drug will ensure that any cue paired with it has its motivational power boosted by a strong dopamine signal. Because the sensitization this produces is stable, the next time the cue is encountered, one will be highly motivated to take the drug, which, in turn, will entrench the sensitization. Drug-taking enhances the motivational power of anything associated with it to motivate drug-taking itself.
This is only a sketch, but it is enough to allow us to outline how one of the major factors in addiction operates at the psychological level. The psychological correlate of this drug-induced neuroplasticity is a complex dispositional state. We could perhaps characterize it as an intrinsic dispositional motivation. It is a standing state of a person induced by long-term drug use which is triggered into an occurrent state by drug-related cues. The occurrent state that is thus triggered is simply an intrinsic motivation to take drugs, but it comes on the scene directly in response to perceiving a cue.
Many philosophers have found it promising to think of the distinctive psychological state in addiction as a desire of some sort. R. Jay Wallace, Gary Watson, Brendan Dill de Kenessey & Richard Holton, and Federico Burdman, among others, have defended this view. I suggest that such theories must be revisited in light of the incentive sensitization theory.
For simplicity, let’s illustrate this with reference to Wallace’s view. Wallace thinks, in general, that motivational states come in two varieties. There are what we might call volitions, on the one hand, and desires proper, on the other. Desires proper are motivational states that are merely given. They have a distinctive phenomenology and are not judgment-sensitive. Natural appetites, such as those for food, drink, and sex, are like this. There is a distinctive, uncomfortable way that it feels to want those things, and this distinctive phenomenology and whatever motivational power it has are impervious to judgment. Such desires can serve as inputs to deliberation, but when one is in a deliberative mode, rarely lead directly to action. Volitions, such as intentions and (the products of) choices are outputs of deliberation and are under the direct control of the will. They also tend to lead to action directly if they are not inhibited.
Not all philosophers are happy to countenance the existence of a will in this sense, but I am sympathetic to Wallace’s general picture. It has a Kantian flavor in that we can be heteronymous with respect to desires proper if we fail to take enough reflective distance from them. It is with this background that Wallace identifies the distinctive motivational state in addiction as a kind of desire proper. We should ask: Given what we have seen from the incentive sensitization theory above, is he right?
There is considerable plausibility to the idea that over the course of addiction, people with addictions develop what is, in effect, an acquired appetite, analogous to the appetites that we have naturally but which are gradually built up over periods of drug use. Once someone has such a standing state installed, they are sometimes subject to an appetite-like craving with urgency and force akin to hunger or thirst. It is also natural that, during periods of heavy drug use, these appetitives have a periodicity analogous to that we find in hunger and thirst, i.e., they are temporarily satisfied by consumption but return with regularity.
So far, this is all quite consistent with the incentive sensitization theory. The neuroplasticity resulting from long-term drug use puts the agent into a stable dispositional state analogous to those corresponding to the natural appetites. But the theory also suggests that the motivation to take drugs should be highly cue-dependent, so, in addition to any periodicity that may accompany a lifestyle of regular drug use, we should expect acute occurrent episodes of wanting to take drugs to occur in response to encountering things associated with drug use. This could explain why, unlike natural appetites, addictive appetites can often seem to have no satisficing point, with addicts often reporting feeling that they can never get enough. As long as the cues are present, the relevant triggering will occur. Compare hunger: a giant platter of cheese-smothered hot garbage, however strongly one might be inclined towards it under other circumstances, may appear highly unattractive immediately after one has just eaten a particularly large meal.
There is also good reason to think that once a drug-related cue is encountered, deliberation is often cut out of the process of determining how the agent will act. Of course, we do sometimes act on natural desires without much deliberation. But deliberation playing less of a role than normal in the face of addictive motivation would go a long way to explaining the puzzle of addiction itself: why do people continue to use despite knowledge of adverse consequences? Richard Holton and Kent Berridge have also noted this in their joint work on addiction: “[A]ddictive desire does not typically function like [ordinary desire]. It does not serve as an input to deliberation, something to be weighed, along with other competing desires, in deciding what to do. Instead, addictive desire functions as something more like an intention: as something that, unless checked, will lead, in a rather direct way, to action.”
I think that the motivational states identified by the incentive sensitization model share too many interesting properties with intentions to be unproblematically considered desires. Following Bratman, we might say that intentions have stability; they preserve the motivational force of a choice or decision for later, resisting revision—and are controlling—unless revised, they lead the agent to act directly. The relevant states in addiction seem to be like that, yet they come unbidden and are not judgment-sensitive. That is, they straddle the boundaries of Wallace’s moral psychology. I, therefore, call them hybrid intentions.
Hybrid intentions are action-controlling because of the properties they share with intentions. Still, it is very difficult for the agent him- or herself to exercise control over them in turn because of their judgment-insensitivity. This makes hybrid intentions interesting states but also means they are liabilities with respect to acting well. Their formation is not subject to volitional control, but they possess a high degree of motivational efficacy that persists over time. Moreover, to the extent that they operate by cutting deliberation out of the process of determining how the agent will act, the agent’s powers of self-control are diminished. So, it isn’t that the agent is led astray by a forceful desire putting its thumb on the scales in deliberation. Deliberation is cut out of the process of determining what will be done.
This is, of course, only a sketch, but it is enough to allow us to begin drawing out the moral psychological import of addiction. For instance, we know that the motivational power of hybrid intentions is considerable. To the extent that this means that action results without any contribution from the agent’s considered judgment, values, or deep self (depending on one’s view of moral responsibility), this could be thought of as a gradable excuse for taking drugs in a situation where the balance of reasons favors not doing so. However closely the force and persistence of the hybrid intention approximates literal compulsion, the stronger the excuse.
It is also clear that whatever the motivational force of a hybrid intention, it doesn’t typically rise to the level of literal compulsion. How else could we explain that addicts respond to incentives or manage to recover? This means, perhaps unsurprisingly, that hybrid intentions do not operate in isolation from other things the agent believes, wants, or is committed to. Further, we also know that those with addictions respond positively to therapeutic support, to well-crafted choice architecture, to the cultivation of self-insight, and perhaps more crucially, to being held accountable to a community. All of this only seems appropriate if we see addiction broadly within the framework of agency. The fact that recovery is possible under such conditions suggests that failure to do so soon enough or effectively enough could render an agent negatively assessable if enough supportive background conditions are satisfied.
A second form of assessment also seems appropriate. The standing state of the agent, which is installed by long-term drug use, the susceptibility to being cued into a hybrid intention, is a state of the agent. Some philosophers believe that we can assess someone’s character—roughly, one’s more-or-less stable set of cognitive, behavioral, and affective dispositions—as such, regardless of provenance. The states in question may be states for which the agent is not fully blameworthy—if, for instance, the gradual building up of the dispositional state was done in ignorance of consequences of drug use— but in general, that is not a condition on being an object of aretaic assessment. As Susan Wolf points out, no one is wholly self-created, but we nevertheless (rightly, it seems) take the ways people are to have great interpersonal significance, which grounds a diverse range of judgments and emotions (as theorists such as Angela Smith emphasize.)
The foregoing presupposes the distinction between responsibility as accountability vs. responsibility as attributability. This is a controversial distinction, but the basic idea, due to Gary Watson, is that one strand in our thinking about moral responsibility tracks avoidability. Blame is an example of holding agents accountable. It is a quasi-punitive social sanction. It is a response to actions and is sensitive to excusing conditions. But another strand tracks agents’ features—what is attributable to them—and does not require avoidability. Having the sort of volitional impairment suggested by the incentive sensitization theory—dispositional as it is—just seems like a non-ideal way to be. It makes it harder, other things being equal, to translate one’s practical judgments into action and, therefore, harder to live according to one’s values. Any inclination we may feel to assess agents negatively for being that way may derive its justification simply from its intrinsic moral importance.
From the point of view of the big picture project, what matters most is that we begin to make progress in understanding how to cash out the personal and interpersonal significance of addiction, even if one rejects some of the substantive value-theoretic assumptions I have adopted. Nevertheless, using the incentive sensitization theory of addiction, I hope to have pointed the way to a number of claims.
First, the pharmacology of addictive drugs and the neurobiology of addiction support the claim that there is a distinctive kind of volitional impairment in addiction. The nature of this impairment means that an analysis of addiction in terms of recalcitrant desires alone is inadequate; the operant motivational states in addiction share too many interesting properties with intentions. Nevertheless, the neuroplasticity characteristic of addiction corresponds to traits of character that are morally assessable. Further, because addiction is consistent with agency, recovery is predicted by treating those with addictions as agents and by providing other scaffolds to self-control, e.g., self-insight, accountability, choice architecture, and other forms of support. Failure to recover under such conditions of support is potentially negatively assessable.
The post Reconciling and Manifest and Scientific Images of Addiction first appeared on Blog of the APA.
Read the full article which is published on APA Online (external link)
