For recovering addicts, the sight of drug-taking paraphernalia and other reminders of drug use can trigger intense cravings and relapses. Now, two studies report that it is possible to impair rats' memories associated with taking cocaine and that such treatments significantly reduce their drug-seeking behaviours.

Both studies are based on the belief that retrieval of some forms of memory are followed by an active 'refiling' process, known as memory reconsolidation. If this process is disrupted, memories might be weakened or even lost.

In the first study, Lee and colleagues trained rats to self-administer cocaine by nosepoking, thereby establishing a strong addiction. The animals then learned the association between nosepoking, cocaine infusion and the illumination of a light. This was followed by a 'reactivation' session, in which nosepoking resulted only in the light coming on. Before this session, the researchers blocked transcription of Zif268 — a gene involved in reconsolidation — by infusing the antisense oligodeoxynucleotides into the amygdala. A few days later, the rats were taken to the same chamber, but this time two levers had been installed in it: pressing one lever lit the light associated with cocaine use, whereas pressing the other did nothing. Rats that had received the antisense treatment pressed the lever that illuminated the light significantly fewer times than their control counterparts.

In the second study, Miller and Marshall used the model of conditioned place preference, in which rats learned to associate the rewarding effects of cocaine with one chamber where the drug was provided, and later preferred to stay in that chamber even when no cocaine was available. Interestingly, the researchers found that this phenomenon was associated with activation of ERK, CREB, ELK1 (a member of the ETS oncogene family) and Fos in the accumbens core, which is involved in the initiation and maintenance of drug-seeking behaviours, but not in the accumbens shell, which mediates the primary rewarding effects of drugs. Infusion of an inhibitor of the ERK pathway, U0126, into the accumbens core blocks the activation of ERK, CREB, ELK1 and Fos as well as the animals' preference for the cocaine chamber.

As only drug-related memories were being recalled when the inhibitors were given, the resulting amnesia might be specific to those memories rather than having a general effect on all memories. These findings hint at an exciting new approach that might help addicts kick the habit.