Opiates suppress the inhibitory effect of GABAergic neurons on dopaminergic neurons in the ventral tegmental area, resulting in an increased activity of DA neurotransmission in the ventral striatum (Kreek 2008). Opiates also strongly impact on opiate receptors and inhibit noradrenergic neurons in the locus coeruleus (Kreek 2008). The thus suppressed noradrenergic activity results in clinically manifest symptoms of opiate intoxication such as bradycardia, drowsiness, and slow respiration. This opioidergic inhibition of noradrenergic neurons in the locus coeruleus results in counteradaptive upregulation of noradrenergic neurotransmission, resulting in a new homeostatic balance between noradrenergic neuroadaptation and opioidergic inhibition as long as opiate intake is maintained on a relatively regular basis. However, when opioidergic inhibition of noradrenergic neurotransmission is suddenly interrupted during abstinence, noradrenergic hyperactivity occurs with typical physical and affective withdrawal symptoms, e.g., tachycardia, transpiration, restlessness, sleep disorders, and anxiety (Kosten and George 2002). Because of heroin’s short half-life period of 6 h, dependent subjects need to consume heroin several times a day in order to avoid aversive withdrawal symptoms, thus negatively reinforcing drug intake via the avoidance of a highly unpleasant state. Interestingly, withdrawal symptoms can also occur as a consequence of context-conditioned learning: Siegel (1983) observed that counteradaptive mechanisms lead to withdrawal symptoms as soon as rats used to opiate application in a certain context (cage) are left in this environment without opiate supply. Context-dependent cues thus seem to activate, e.g., noradrenergic upregulation, resulting in hyper-excitation once opiate application does not occur as expected. If on the other hand usual opiate doses were applied in an unfamiliar context, e.g., an unknown cage, animals used to a certain opiate dose displayed symptoms of opiate intoxication and even died, indicating that in the unknown context, their organisms failed to actively prepare for renewed intoxication via context-conditioned neuroadaptational processes (Siegel 1983). In humans, comparable conditioned effects of chronic drug intake may manifest as perception biases towards drug-associated stimuli and memory including related emotions and modification of (aversive) emotional states. For instance, Lubman and colleagues observed that long-term heroin users vs. healthy controls displayed altered processing of drug-related (e.g., pictures of drug preparation and injection) and pleasant pictures (e.g., pictures of food, erotic nudes): positive pictures, contrary to drug-related pictures, elicited less physiological responses and lower self-reported arousal in drug-dependent subjects. Further, low valence ratings of pleasant pictures predicted consistently regular heroin use (at least once a week) at a median of a 6-month follow-up assessment (Lubman et al. 2009). Like in alcohol dependence, prolonged abstinence allows for neurofunctional recovery, hypothetically reflected in decreased salience of drug-associated cues, thereby, reducing the risk of relapses. Indeed, an fMRI study indicated that following long-term compared to short heroin abstinence, subjects displayed decreased neural responses to heroin-related cues in brain regions subserving visual sensory processing, attention, memory, and action planning (Lou et al. 2012).

The strong effects of cocaine on the DA system can be demonstrated by the fact that the consumption of this substance discharges about 400 % of the dopamine doses normally released by natural reinforcers, such as food or sexual activities (Hurd et al. 1990). These massive effects lead to a counteradaptive reduction in (striatal) DA D₂ receptor availability in cocaine-dependent subjects (Volkow et al. 1993), and this striatal downregulation predicted reduced functional responses to a non-drug-related reinforcer (monetary reward) in the thalamus, a brain region that is implicated in DA-modulated conditioned responses and reward expectation (Asensio et al. 2010). However, in this study also a low DA D₂ receptor availability in the dorsal and ventral striatum was found to be correlated with high responsiveness of the dorsal superior medial frontal gyrus (Brodmann area 6/8/32), an area that is crucially involved in behavioral monitoring. Asensio and coworkers assumed that cocaine-addicted individuals who displayed lower ventral striatal DA D₂ receptor availability are also the ones who possibly need to compensate particularly for their reduced sensitivity to monetary reinforcement (Goldstein et al. 2007) or for their increased impulsivity (Asensio et al. 2010). Other fMRI-based study revealed reduced brain activation elicited by naturally reinforcing visual stimuli, e.g., pictures with erotic content, while cocaine-associated cues which triggered craving elicited increased functional responses throughout the brain including prefrontal, limbic, and parietal regions (Garavan et al. 2000). It has been hypothesized that prefrontal brain activation elicited by drug-related stimuli impairs executive function and behavior control (Beck et al. 2012). Indeed, a prospective fMRI study by Paulus et al. (2005) observed that decreased activation of the PFC during a decision-making task was associated with the subsequent relapse risk in methamphetamine-dependent patients.