In their target article, Huang & Bargh (H&B) suggest an interesting comparison between genes and goals. Inspired from the selfish gene theory, which proposes that genes are in competition for survival and reproduction and that the phenotypic organism only serves as a “survival machine” (Dawkins Reference Dawkins1976), they inquire in how far a similar framework might work for behavioral goals. The starting point for this argument appears in the observation that humans can pursue different goals. Because we cannot observe “goals” directly, we have to infer them from observing goal-directed behaviors. Individual goals develop and try to get access to behavioral output, which results in a behavioral phenotype. Importantly, there is usually only one behavior that can be emitted at a time, no matter how many goals are established in the organism. It appears all too obvious that there must be some sort of competition between distinct goals for access to behavioral output, in particular because single goals can be counteracting and resulting in behaviors that rule out one another. This goal competition was now taken to a new level by H&B by assigning every goal an innate “selfishness.” Thus, goals are selfish and “fight” for their access to behavioral output. H&B argue that this competition is the reason for often inconsistent behaviors with highly disadvantageous components for the individual.

Here, we ask whether the selfish goal concept may well be suitable not only to explain inconsistencies in micro-behaviors, but also in the gross behavioral repertoire of an individual. This is often (but not always) associated with psychopathologies. In a human being with a “normal” behavioral repertoire, one can observe a certain number of goals acting with a balanced influence on behavioral output. This results in a behavioral phenotype that displays a number of distinct goal-directed behaviors. The frequency of each of them depends on whether they are rewarded or punished by the environment. Thereby, an increase in the frequency of one behavior usually means a decrease in the frequency of other behaviors, which may be seen as a behavioral competition. This competition of behaviors may directly reflect a competition of the goals at neuronal level in the brain. Although a majority of humans can maintain a certain amount of rewarded goal-directed behaviors, for some individuals this balance becomes unusually polarized. In that case one “very selfish” goal has taken control over the behavioral repertoire. Likewise, at neuronal level one goal representation might inhibit all others in a “collateral inhibition” for access to behavioral output. This imbalance may occasionally result in a “successful behavioral phenotype,” for example, in a professional tennis player, who is playing tennis all the time wishing to win as many tournaments as possible, but suppressing most other behavioral alternatives. Quite often, however, selfish goal dominance may result in psychopathologies, such as drug addiction.

Drug addiction is a syndrome in which the seeking and taking of psychoactive drugs has gained higher priority than virtually all other normally important behaviors. In that, drug addiction is a good example of how mostly unconscious goals, which were developed in instrumental and classical conditioning processes (Robbins et al. Reference Robbins, Ersche and Everitt2008; White Reference White1996), may dominate conscious goals of becoming abstinent and spending more time with other activities. For drug addiction, there is now a fairly good understanding of how subconscious goals dominate cognitive ones at neurophysiological level. This may, in turn, inform theories of normal goal competition. Addictive drugs can induce a learning-like neuronal plasticity already after a single administration at synapses of the mesolimbic reward system (Saal et al. Reference Saal, Dong, Bonci and Malenka2003; Ungless et al. Reference Ungless, Whistler, Malenka and Bonci2001). During repeated drug administration, classical and instrumental learning processes for goal-directed behaviors are established. Thus, stimulus (S)-response (R) associations are formed, which establish the goal of “mental state change” by means of a drug and further elaborated drug use mechanisms (Müller & Schumann Reference Müller and Schumann2011).

However, conscious goal representations may also be formed after only a single drug exposition, by establishing an episodic memory of a desirable drug-induced mental state (Müller Reference Müller2013). At physiological level, there is an increase in dendritic branching and synaptic spine density in neurons of the reward circuitry, but also in neocortical structures (Robinson & Kolb Reference Robinson and Kolb2004). Importantly, this drug-induced neuronal plasticity prevents establishment of neuronal plasticity induced by natural stimuli learning (Kolb et al. Reference Kolb, Gorny, Li, Samaha and Robinson2003). Here one may speculate whether this may be one site where drug-goals become “selfish,” that is, by suppressing establishment of plasticity and learning directed toward alternative reinforces. Drug addiction requires long-term drug consumption with an escalation of the consumption and a loss of control over this behavior (American Psychiatric Association 1994). The loss of control may be seen as a goal conflict or as an imposition of one goal over the other in which the unconscious goal of “harm avoidance” and the conscious goal of “abstinence” lose influence on the unconscious goal of “drug taking” and the conscious goal of “changing the state of mind” (Müller & Schumann Reference Müller and Schumann2011).

Why do these goals lose out? At physiological level, there is evidence that the initially outcome-controlled behavior of drug taking eventually becomes a cue-controlled habit. It is no longer dependent on the actual effects of the drugs, but depends mainly on drug predicting stimuli. This was shown to be mediated by a neuroanatomical loop projection from the nucleus accumbens to the ventral tegmental area and from there to the dorsal striatum (Belin & Everitt Reference Belin and Everitt2008; Haber et al. Reference Haber, Fudge and McFarland2000). Habits are controlled by the dorsal striatum and are difficult to suppress by neocortical inhibitory projections (Knowlton et al. Reference Knowlton, Mangels and Squire1996). Furthermore, in drug addiction, prefrontal function and behavioral control decline steadily, possibly by a reduction of its serotonergic and dopaminergic innervation (Müller et al. Reference Müller, Pum, Schumann, Huston, Müller and Jacobs2010; Pelloux et al. Reference Pelloux, Dilleen, Economidou, Theobald and Everitt2012). This reduces prefrontal inhibition of striatal circuits and allows for a more compulsive pursuit of habit-like behaviors. In other words, cortically represented goals are weakened and lose their influence on behavioral output while striatally represented goals become superdominant. Altogether, we suggest that these pathways (Belin et al. Reference Belin, Jonkman, Dickinson, Robbins and Everitt2009) may serve as one physiological base for H&B's conscious and unconscious competition of goals that appear indeed “selfish” when their impact on behavioral output is considered.