Addiction,
craving, and control
Why does drug use persist
despite serious costs?
Hanna Pickard · Introduction and chapter 6
The addiction unit
Addiction,
PHL/NEU 251 · Samuel Murray
01A fictional case
Jordan’s drinking
Jordan’s drinking is damaging their health and work.
Jordan understands these costs and wants to stop.
On repeated occasions, Jordan nevertheless drinks.
Why?
Pickard’s starting point
What needs explaining?
Drug use persists despite evident, severe costs to the person’s own good.
| Craving | A desire to use. |
| Physical dependence | Adaptation to a drug; stopping can produce withdrawal. |
| Addiction | The persistent, seriously costly pattern we are trying to explain. |
Pickard, Introduction, pp. 10–11. Her account is paraphrased here.
03The person and their circumstances
Pickard’s approach to explanation
Psychology first
What does the drug do for this person?
Reasons, feelings, history, and circumstances.
Heterogeneity
What makes this person’s situation different?
Similar patterns of use can have different explanations.
Psychological explanation can incorporate neuroscience. Some cases may involve brain pathology.
Pickard, Introduction, pp. 6–11.
04Motivation and enjoyment
Wanting and liking
Wanting
Motivation to obtain something.
Liking
Its pleasurable impact.
A rat experiment
Amphetamine in the nucleus accumbens increased seeking for sucrose when a learned cue appeared.
A separate measure
Positive taste reactions did not increase.
Would less enjoyment necessarily mean less pursuit?
PCN, pp. 471–472; Wyvell & Berridge (2000).
05Learning and motivation
How drug use becomes reinforced
A valued effect
A pleasurable effect can make use more likely to recur.
Relief
Reducing an unpleasant state can also reinforce use.
Learned cues
A setting associated with drug effects can acquire motivational significance.
What would show that these processes impair intentional action?
PCN, pp. 470–474. Connection to Session 6’s discussion of agency.
06Development, exposure, and recovery
Age and vulnerability
Vulnerability
Earlier use is associated with greater risk for some substance use disorders.
Development and circumstances both matter. Adults remain vulnerable.
Recovery over time
Many people remit. Some problems persist or recur.
“Maturing out” does not mean automatic recovery at 30.
What could age restrictions do?
May, p. 130; Volkow et al. (2021); Lopez-Quintero et al. (2011); McCabe et al. (2022).
07Pain relief, risk, and access to care
Opioid prescribing and prevention
Why permit a treatment that carries a risk of addiction?
Legal permission
U.S. federal law requires a legitimate medical purpose within professional practice.
Clinical judgment
Weigh pain relief against risks. Consider alternatives, assess risk, and follow up.
CDC prescribing guidance is not itself a law.
What evidence would show that a stricter rule reduces harm while preserving needed care?
Count health outcomes and access to care, as well as prescriptions.
21 CFR §1306.04(a); CDC prescribing guidance. U.S. federal baseline; state rules vary.
08The strong compulsion account
Irresistible craving
Use persists because the desire for drugs cannot be resisted.
What would distinguish impossible to resist from very difficult to resist?
Pickard’s test: do drugs invariably prevail when alternatives are available?
Pickard, pp. 81–84.
07Competing explanations
Habit and goal-directed action
Habitual responding
A learned stimulus–response relation guides behavior.
Relatively insensitive to the outcome’s current value.
Goal-directed action
An expected outcome and its current value guide behavior.
Sensitive to what the action will achieve.
Disease is a classification. Habit is a proposed mechanism.
“Goal-directed” here does not mean wise or consciously deliberated.
Pickard, pp. 84–88. Habit is used here in its technical learning-theory sense.
08Outcome devaluation · schematic procedure
Testing an expected outcome
01
Learn
Pressing a lever delivers a particular food.
02
Devalue
Let the animal eat its fill of that food.
03
Test
Compare pressing with a control group. Test presses deliver no food.
What would each explanation predict?
Less pressing
Supports sensitivity to the outcome’s current value.
Unchanged pressing
Can suggest habit, but requires further checks.
Pickard, pp. 84–87. A teaching illustration of the procedure, not a new dataset.
09Animal choice · prediction
Cocaine and an alternative reward
Lever C
Cocaine
Intravenous delivery
Lever S
No reward
At first
What do you predict will happen
to their preference?
to their preference?
First, the rats sample the available outcomes.
Lenoir et al. (2007), Figure 5a · schematic animation.
10Animal choice · results
Preference changes with the alternative
What happens when
the alternative changes?
the alternative changes?
The vertical axisPositive values favor Lever S.
Negative values favor Lever C.
Zero indicates no preference.
The changeLever S supplies saccharin from day 1. Cocaine remains available.
The findingThe group shifts toward the sweet alternative.
This shows flexibility under these conditions.
Lenoir et al. (2007), Figure 5a (cropped) · n = 24 · published error bars · CC BY.
11A randomized human study
Incentives in human treatment
40 adults received community-reinforcement treatment.
One group also received vouchers for cocaine-negative urine tests.
Vouchers operated during weeks 1–12. Both groups were treated alike in weeks 13–24.
What does this show about changing the conditions for recovery?
Higgins et al. (1994) · chart of reported group means; error bars omitted.
12The philosophical inference
What follows about control?
Behavior can remain responsive to alternatives.
What control exists in the person’s actual circumstances?
How difficult is it to exercise that control?
Can the person sustain change over time?
Could repeated difficulty explain persistence even without absolute irresistibility?
Pickard, pp. 89–92. The last question is an objection for discussion.
13Return to the fictional case · pair discussion
Jordan’s available alternatives
In both versions, Jordan understands the harm and wants to change.
Dependable support
Appointments are reachable.
Valued activities are accessible.
Someone reliably offers help.
Limited access
Services are hard to reach.
Companionship centers on drinking.
Offers of help are unreliable.
What changes about the difficulty of acting otherwise?
Propose one change that would make a valued alternative more accessible. Explain why it might help.
Original comparison applying Pickard’s approach. These circumstances do not determine either outcome.
14Responsibility for a particular action
Knowledge, control, and responsibility
Further hypothetical: Jordan misses a work obligation after drinking.
Knowledge
What did Jordan understand?
Control
What could Jordan do in these circumstances?
The particular action
What was harmful or wrong?
An appropriate response
What response is justified?
Does recognizing some agency commit us to blame or to saying “try harder”?
Responsibility for drinking and for associated conduct require separate assessment.
Pickard, Introduction, pp. 11–17. A preview of the later responsibility-without-blame discussion.
15A short written response
What the evidence leaves open
Choose one finding from today.
Explain what it shows about addiction.
Identify one question it leaves unsettled
about control or responsibility.
Next: responsibility, blame, and relationships.
16Readings and evidence
Sources
Hanna Pickard (2026)
What Would You Do Alone in a Cage with Nothing but Cocaine?
Introduction, pp. 1–20; chapter 6, pp. 81–92.
Princeton University Press.
Purves et al. (2013)
Principles of Cognitive Neuroscience, 2nd ed.
Chapter 14, pp. 470–476.
Background on reward and learning.
Wyvell & Berridge (2000)
Intra-accumbens amphetamine increases the conditioned incentive salience of sucrose reward.
Journal of Neuroscience 20: 8122–8130.
Lenoir et al. (2007)
Intense sweetness surpasses cocaine reward.
PLoS ONE 2: e698.
Figure 5a reproduced under CC BY.
Higgins et al. (1994)
Incentives improve outcome in outpatient behavioral treatment of cocaine dependence.
Archives of General Psychiatry 51: 568–576.
Class discussion
Questions adapted anonymously from student submissions. Jordan and the comparison of circumstances are fictional teaching examples.
Study links open externally. The presentation and its figures work offline.
17Further evidence and guidance
Age and prescribing: sources
Volkow et al. (2021)
Substance use disorders by time since first use.
JAMA Pediatrics 175: 640–643.
Prescription-drug analyses concern misuse.
Lopez-Quintero et al. (2011)
Probability and predictors of remission.
Addiction 106: 657–669.
Timelines are measured since onset.
McCabe et al. (2022)
Symptoms at age 18 and substance use disorder in adulthood.
JAMA Network Open 5: e225324.
Sartor et al. (2009)
Timing of first alcohol use and alcohol dependence: common genetic influences.
Addiction 104: 1512–1518.
CDC Clinical Practice Guideline (2022)
Opioids for outpatient pain care in adults.
Scope and exceptions matter.
Current recommendations and principles
21 CFR §1306.04(a)
Purpose of controlled-substance prescriptions.
DEA prescriptions guidance
Federal baseline checked September 2026.
Policy implications are questions for evaluation; these studies do not establish the effects of a particular age ban.
20