[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Original Investigation
November 2016

Incubation of Cue-Induced Craving in Adults Addicted to Cocaine Measured by Electroencephalography

Author Affiliations
  • 1Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York
  • 2Department of Neuroscience, Icahn School of Medicine at Mount Sinai, New York, New York
JAMA Psychiatry. 2016;73(11):1127-1134. doi:10.1001/jamapsychiatry.2016.2181
Key Points

Question  Does cue-induced craving follow a parabolic trajectory as a function of duration of abstinence in humans with cocaine use disorders, as previously shown in preclinical studies?

Findings  The results of this laboratory-based cross-sectional study confirm the prior preclinical reports of a parabolic trajectory of cue-induced craving, but only when assessed objectively using the electroencephalographic measures of cue reactivity and not via subjective reports of craving.

Meaning  These translational findings highlight an abstinence period during which individuals addicted to cocaine are perhaps most vulnerable to relapse and show that the use of objective assessments is important to uncover such a clinically crucial phenomenon in individuals addicted to cocaine who are seeking treatment.


Importance  A common trigger for relapse in drug addiction is the experience of craving via exposure to cues previously associated with drug use. Preclinical studies have consistently demonstrated incubation of cue-induced drug-seeking during the initial phase of abstinence, followed by a decline over time. In humans, the incubation effect has been shown for alcohol, nicotine, and methamphetamine addictions, but not for heroin or cocaine addiction. Understanding the trajectory of cue-induced craving during abstinence in humans is of importance for addiction medicine.

Objective  To assess cue-induced craving for cocaine in humans using both subjective and objective indices of cue-elicited responses.

Design, Setting, and Participants  Seventy-six individuals addicted to cocaine with varying durations of abstinence (ie, 2 days, 1 week, 1 month, 6 months, and 1 year) participated in this laboratory-based cross-sectional study from June 19, 2007, to November 26, 2012. The late positive potential component of electroencephalography, a recognized marker of incentive salience, was used to track motivated attention to drug cues across these self-selected groups. Participants also completed subjective ratings of craving for cocaine before presentation of a cue, and ratings of cocaine “liking” (hedonic feelings toward cocaine) and “wanting” (craving for cocaine) after presentation of cocaine-related pictures. Data analysis was conducted from June 5, 2015, to March 30, 2016.

Main Outcomes and Measures  The late positive potential amplitudes and ratings of liking and wanting cocaine in response to cocaine-related pictures were quantified and compared across groups.

Results  Among the 76 individuals addicted to cocaine, 19 (25%) were abstinent for 2 days, 20 (26%) were abstinent for 1 week, 15 (20%) were abstinent for 1 month, 12 (16%) were abstinent for 6 months, and 10 (13%) were abstinent for 1 year. In response to drug cues, the mean (SD) late positive potential amplitudes showed a parabolic trajectory that was higher at 1 (1.26 [1.36] µV) and 6 (1.17 [1.19] µV) months of abstinence and lower at 2 days (0.17 [1.09] µV), 1 week (0.36 [1.26] µV), and 1 year (–0.27 [1.74] µV) of abstinence (P = .02, partial η2 = 0.16). In contrast, the subjective assessment of baseline craving (mean [SD] rating: 2 days, 26.05 [9.85]; 1 week, 18.70 [11.01]; 1 month, 10.87 [10.70]; 6 months, 6.92 [8.47]; and 1 year, 3.00 [3.77]) and cue-induced liking (mean [SD] rating: 2 days, 3.06 [2.34]; 1 week, 2.33 [2.87]; 1 month, 1.15 [2.03]; 6 months, 1.00 [2.24]; and 1 year, 1.00 [1.26]) and wanting (mean [SD] rating: 2 days, 3.44 [2.62]; 1 week, 2.72 [2.87]; 1 month, 1.46 [2.33]; 6 months, 1.00 [2.16]; and 1 year, 1.00 [1.55]) of cocaine showed a linear decline from 2 days to 1 year of abstinence (P ≤ .001, partial η2 > 0.26).

Conclusions and Relevance  The late positive potential responses to drug cues, indicative of motivated attention, showed a trajectory similar to that reported in animal models. In contrast, we did not detect incubation of subjective cue-induced craving. Thus, the objective electroencephalographic measure may possibly be a better indicator of vulnerability to cue-induced relapse than subjective reports of craving, although this hypothesis must be empirically tested. These results suggest the importance of deploying intervention between 1 month and 6 months of abstinence, when addicted individuals may be most vulnerable to, and perhaps least cognizant of, risk of relapse.