Brain activation covaries with reported criminal behaviors when making risky choices: A fuzzy-trace theory approach

Criminal behavior has been associated with abnormal neural activity when people experience risks and rewards or exercise inhibition. However, neural substrates of mental representations that underlie criminal and noncriminal risk-taking in adulthood have received scant attention. We take a new approach, applying fuzzy-trace theory, to examine neural substrates of risk preferences and criminality. We extend ideas about gist (simple meaning) and verbatim (precise risk-reward tradeoffs) representations used to explain adolescent risk-taking to uncover neural correlates of developmentally inappropriate adult risk-taking. We tested predictions using a risky-choice framing task completed in the MRI scanner, and examined neural covariation with self-reported criminal and noncriminal risk-taking. As predicted, risk-taking was correlated with a behavioral pattern of risk preferences called “reverse framing” (preferring sure losses over a risky option and a risky option over sure gains, the opposite of typical framing biases) that has been linked to risky behavior in adolescents and is rarely observed in nondisordered adults. Experimental manipulations confirmed processing interpretations of typical framing (gist-based) and reverse-framing (verbatim-based) risk preferences. In the brain, covariation with criminal and noncriminal risk-taking was observed predominantly when subjects made reverse-framing choices. Noncriminal risk-taking behavior was associated with emotional reactivity (amygdala) and reward motivation (striatal) areas, whereas criminal behavior was associated with greater activation in temporal and parietal cortices, their junction, and insula. When subjects made more developmentally typical framing choices, reflecting non-preferred gist processing, activation in dorsolateral prefrontal cortex covaried with criminal risk-taking, which may reflect cognitive effort to process gist while inhibiting preferred verbatim processing