The noradrenergic system modulates performance on tasks dependent on semantic and associative network flexibility (NF) in individuals without neurodevelopmental diagnoses in experiments using a beta-adrenergic antagonist, propranolol. Some studies suggest drugs decreasing noradrenergic activity are beneficial in ASD. In individuals without neurodevelopmental diagnoses, propranolol is beneficial only for difficult NF-dependent problems. However, in populations with altered noradrenergic regulation, propranolol also benefits performance for simple problems. Due to decreased flexibility of access to networks in ASD, we wished to examine the effect of propranolol on NF in ASD. ASD subjects benefited from propranolol on simple anagrams, whereas control subjects were impaired by propranolol. Further study will be necessary to confirm this finding in a larger sample and to compare clinical response with cognitive response to propranolol.

Norepinephrine and dopamine are both believed to affect signal-to-noise in the cerebral cortex. Dopaminergic agents appear to modulate semantic networks during indirect semantic priming, but do not appear to affect problem solving dependent on access to semantic networks. Noradrenergic agents, though, do affect semantic network dependent problem solving. We wished to examine whether noradrenergic agents affect indirect semantic priming. Subjects attended three sessions: one each after propranolol (40 mg) (noradrenergic antagonist), ephedrine (25 mg) (noradrenergic agonist), and placebo. During each session, closely related, distantly related, and unrelated pairs were presented. Reaction times for a lexical decision task on the target words (second word in the pair) were recorded. No decrease in indirect semantic priming occurred with ephedrine. Furthermore, across all three drugs, a main effect of semantic relatedness was found, but no main effect of drug, and no drug/semantic relatedness interaction effect. These findings suggest that noradrenergic agents, with these drugs and at these doses, do not affect indirect semantic priming with the potency of dopaminergic drugs at the doses previously studied. In the context of this previous work, this suggests that more automatic processes such as priming and more controlled searches of the lexical and semantic networks such as problem solving may be mediated, at least in part, by distinct mechanisms with differing effects of pharmacological modulation.