The brain's decision making circuitry is strongly affected by drugs of abuse and this can give rise to decisions that perpetuate addictive behavior. These changes however may be malleable. Existing addiction treatments seem to restore some decision processes to a healthier state by preferentially changing activity in value and decision making brain circuits. The goal of our work is to understand how addiction and addiction treatment may impact the neural computations underlying decision making in a dynamic way. Using fMRI, computational modeling, and choice experiments in patients with opioid use disorder, our aim is to bring decision neuroscience to the clinic to make accurate predictions about an individual's likely trajectory in recovery. We are also working to map the ‘decision parameter space’ of a pro-addiction phenotype in subclinical populations.

In parallel, we are developing paradigms to measure the influence of motivational states on decision making. Here, we combine elicitation techniques that causally change a person’s e.g., desire, anxiety, stress level and measure the specific ways these states shift features of the decision process. We are also interested in how these states evolve more naturalistically using experience sampling in the real world. Some questions we are working to answer include: Is craving (specific desire) a gain control-like scaling of subjective valuation? Why does everything else seem insignificant when desire is high? Does craving constrain behavior more compared with hunger and other more global affective states like anxiety or stress? How wide are generalization gradients for the things we like a lot? Why does it feel like there are no substitutes for these things compared with others that we like less? What is the role of the brain's valuation circuit in changing appetitive associative values? Does this circuit update appetitive value or does it work to inhibit its expression?

Most theories of decision making suggest that value is inherently subjective. Multi-attribute utility theories extend this to how these subjective values are constructed and propose that the value of a given course of action or choice option is determined by a weighted sum of its attributes. For example, how much you value a chocolate bar (how satisfied or happy you would be having the bar) depends on how sugary, milky, expensive, and so on that bar is. Furthermore, how much you care about each of these features (attributes), i.e., how much weight you assign to each, will further influence the overall value. If you put a positive weight on milkiness, you might prefer a milk chocolate bar to a dark chocolate bar. And if you are on a diet, you might weigh sugariness more negatively. This can be applied to all types of decisions beyond dietary choice, for example to drug taking and financial decisions. In the lab, we are working to understand what the relevant attributes people consider in a given decision context are and individual differences in how these attributes are weighted. We are also studying how this is implemented in the brain using fMRI. We hypothesize that specific psychiatric symptoms and motivational states can selectively change the weight of some attributes in favor of others (a shift in those representing costs vs. benefits, for example).