Systemic physiological “noise” in fMRI has clinical relevance

Functional magnetic resonance imaging (fMRI) is central to studying neurobiological mechanisms, yet fMRI has limited clinical utility, highlighting the need for novel approaches. We show that a component of the fMRI signal—the systemic low-frequency oscillation (sLFO), linked to blood flow and physiological measures of arousal—indexes trait- and state-level drug use phenotypes. In individuals who chronically use nicotine, sLFO amplitude increased during abstinence and correlated with heightened dependence severity and cue-induced craving. In healthy participants, acute methylphenidate—but not nicotine—reduced sLFO amplitude in a manner that corresponded with improved behavioral performance. These findings demonstrate that the sLFO, typically treated as noise, carries biologically meaningful information. Evaluating the sLFO offers a complementary perspective to traditional fMRI analyses, thus enhancing clinical relevance. Broadly, the sensitivity of sLFO signals to drug administration, cues, and abstinence underscores the need to account for this signal’s contribution when interpreting fMRI responses across experimental conditions.