The hypothesis that serotonin (5-HT) is critically involved in impulse control has been gathering momentum for over 20 years. Soubrié proposed that a common basis for a number of different behavioral effects associated with decreases in brain 5-HT levels was the disinhibition of behavior, which can be directly related to the construct of impulsivity (Soubrie, 1986
). Furthermore, Linnoila and colleagues established a correlation between low levels of the 5-HT metabolite 5-hydroxyindoleacetic acid (5-HIAA) in the cerebrospinal fluid (CSF), and impulsive rather than non-impulsive aggression (Linnoila et al., 1983
). Despite these data indicating a strong association between 5-HT and impulsivity, the extent to which the 5-HT system is involved in ADHD is open to debate. Although low levels of 5-HIAA were observed in the CSF of a group of children and adolescents with disruptive behavioral disorders, including ADHD, this was found to correlate with levels of aggression rather than impulsivity per se, and could reflect the more general role of 5-HT in aggressive behavior (see Miczek, Fish, de Bold, & de Almeida, 2002
for review). However, a trend has also been observed between low levels of 5-HT in the blood and the severity of ADHD symptoms (Spivak et al., 1999
), indicating that this neurotransmitter could somehow be involved in the behavioral disturbances central to this disorder.
In parallel with the dopamine hypothesis, genetic studies indicate an association between ADHD and a polymorphism in the 5-HT transporter gene (Retz, Thome, Blocher, Baader, & Rosler, 2002
), and an allelic variant of the gene encoding the 5-HT1B
receptor (Quist et al., 2003
). Nevertheless, the finding that serotonin-specific re-uptake inhibitors (SSRIs), which increase central 5-HT levels, have little or no effect in ADHD suggests that 5-HT may not be central to the aetiology of the disorder. However, in terms of other impulse-control disorders such as pathological gambling, sexual addiction and personality disorders, SSRIs have proved to offer some therapeutic benefit (Hollander & Rosen, 2000
The effects of acutely reducing levels of 5-HT in the brain through tryptophan depletion on tests of impulsivity and behavioral disinhibition have been investigated in healthy volunteers as well as clinical populations and those with a history of psychiatric disorders. In general tryptophan depletion has been found to increase impulsive responding, particularly in those with a family history of psychiatric disorders, as measured by tests of impulsive action such as the stop-task procedure, go/no-go tasks and the CPT (Crean, Richards, & de Wit, 2002; LeMarquand, Benkelfat, Pihl, Palmour, & Young, 1999; LeMarquand et al., 1998; Quintin et al., 2001; Walderhaug et al., 2002
), yet decreasing CNS 5-HT in this way has no effect in tests of impulsive choice such as delay-discounting and a simple test of probabilistic choice (Anderson, Richell, & Bradshaw, 2003
). However, tryptophan depletion impairs reversal learning in an attentional set shifting task (Rogers et al., 1999
) and subtly alters performance on a gambling test such that subjects did not change as much as controls in response to the size of the reward they could earn (Rogers et al., 2003
). These data suggest that decreasing 5-HT does not necessarily increase impulsive choice per se, but does alter decision-making based on changes in the value of reward and rewarding stimuli.
In the rat, the hypothesis that decreasing 5-HT increases impulsive action has received widespread support. For example, globally reducing forebrain 5-HT through intracerebroventricular (ICV) infusions of the serotonergic toxin 5,7-dihydroxytryptamine, has been shown to increase premature responding on the 5CSRT and profoundly disrupted acquisition and performance of a go/no-go task (Harrison et al., 1997, 1999
). However, whether decreasing 5-HT likewise increases impulsive choice in rats is less clear, as lesions to the serotonergic system have been reported to both increase impulsivity and to have no effect on performance of a delay-discounting paradigm (Bizot, Le Bihan, Puech, Hamon, & Thiebot, 1999; Mobini et al., 2000; Winstanley et al., 2003, 2004; Wogar et al., 1993
). Although the reasons for these discrepancies are unclear, there are a number of obvious differences between these studies, not least the use of different behavioral tasks, varying methodology, and the presence or absence of neuroprotective pre-treatment strategies prior to surgery.
However, not all studies have found an inverse relationship between 5-HT levels and impulsive action. Focusing on the 5CSRT, a recent in vivo microdialysis study using a simplified “one-choice” version of the task found that, although levels of 5-HT did not alter during performance of the task, animals that made more premature responses on the task had higher
levels of 5-HT in the medial PFC (Dalley, Theobald, Eagle, Passetti, & Robbins, 2002
). A positive correlation has also been observed between levels of 5-HT in the right medial PFC and levels of premature responses made on the 5CSRT according to post mortem tissue analysis of trained animals (Puumala & Sirvio, 1998
). It is important to remember that the 5-HT system is highly complex, and over 14 different types of 5-HT receptor have currently been identified which can have both excitatory and inhibitory effects on serotonergic neurons as well as the cells they target, including dopaminergic and glutamatergic neurons. Recent evidence suggests that drugs selective for these different 5-HT receptors can have radically different effects on impulsive behavior (Evenden & Ryan, 1999
). In particular, whereas antagonism of the 5-HT2C
receptor antagonist SB 242084 increases premature responding, the 5-HT2A
receptor antagonist M100907 decreases this measure of impulsivity (Higgins, Enderlin, Haman, & Fletcher, 2003; Winstanley et al., 2003
). Hence, antagonism of certain 5-HT receptors can decrease impulsive responding whereas global 5-HT depletion increases impulsivity on the same behavioral measure.
The essential diversity of the 5-HT system may provide an explanation for the inconsistencies in the literature regarding the role of this neurotransmitter in ADHD and impulsivity. It is clearly oversimplistic to view the serotonergic system as a homogeneous substrate upon which pharmacological agents act to produce monotonic changes in behavior. The finding that different serotonergic manipulations can have contrasting effects on different measures of impulsivity is an important consideration when evaluating the effects of serotonergic drugs in ADHD.