Opioid drugs are essential medications for the relief of serious pain, with no substitutes currently available for postsurgical and other severe indications. Long term use of opioids, however, leads to numerous side effects, and to substantial risk of substance use disorder (SUD). SUD or “addiction” is diagnosed based on behavioral characteristics that manifest broadly as loss of control or “compulsive” drug seeking and impaired decision making or “cognitive flexibility” even after months to years of abstinence. However, the majority of preclinical research of drug abuse focuses on models of drug-taking and reward-seeking rather than on the long-lasting changes in behavioral flexibility that underlie human SUDs. In addition, preclinical studies of SUD mechanism have been limited to comparing animals that have or have not taken drug. This has made it difficult to dissociate opioid- induced changes in biology and behavior that occur merely due to drug exposure from those that actually underlie the pathology of a SUD. We have developed a unique tool to circumvent this significant confound in opioid abuse research. Specifically, we have developed a knock-in mouse that expresses a modified mu opioid receptor (MOR) with altered signaling properties. The MOR when activated by its endogenous ligands, endorphins and enkephalins, engages G protein signaling to control neuronal activity. Following G protein activation by endogenous ligand, most G protein coupled receptors (GPCR), including the MOR, then rapidly recruit arrestins that silence the G protein signal and promote receptor endocytosis and, for the MOR, rapid recycling. This mechanism thereby carefully titrates G protein signal with a precision and time course ideally suited to respond to transmitters that are released in a pulsatile manner. In contrast, MORs activated by morphine and all its derivates effectively engage G protein signaling but poorly engage arrestins. In the current vernacular of GPCR pharmacology, morphine is termed a “biased” agonist, signaling preferentially to G protein over arrestins while endorphins are “balanced” agonists, engaging both G proteins and arrestins. The RMOR receptor was engineered to effectively engage both G protein and arrestin when activated by morphine without altering signaling in response to endogenous transmitters. Importantly, in our extensive previous work, we have found that RMOR mice do not develop tolerance or dependence to morphine nor do they transition to compulsive drug seeking in a model of SUD under conditions where wild type (WT) mice do. More recently, we have found while morphine causes long-lasting changes in cognitive flexibility in WT mice, this effect is also absent in RMOR mice. Here we will use WT and RMOR mice to pinpoint molecular and synaptic mechanisms that underlie SUDs in a paradigm where all mice receive drug but only WT show pathologic morphine responses. We propose that any morphine-induced changes that occur in both genotyp...