Individuals with attention-deficit/hyperactivity disorder (ADHD) exhibit deficits in reward-based learning, which have important implications for behavioral regulation. Prior research has shown that these individuals show altered patterns of risky decision-making, which may be partially explained as a function of dysfunctional reactivity to rewards and punishments. However, research findings on the relationships between ADHD and punishment sensitivity have been mixed. The current study used the Balloon Analog Risk Task (BART) to examine risky decision-making in adults with and without ADHD, with a particular interest in characterizing the manner in which participants react to loss.

612 individuals (Mage = 31.04, SDage = 78.77; 329 females, 283 males) were recruited through the UCLA Consortium for Neuropsychiatric Phenomics (CNP). All participants were administered the Structured Clinical Interview for DSM-IV-TR (SCID-IV), which provided diagnoses used for group comparisons between adults with ADHD (n = 35) and healthy controls (n = 577). A computerized BART paradigm was used to examine impulsivity and risky decision-making, while participants also completed the Barratt Impulsiveness Scale (BIS-11), and ADHD participants completed the Adult Self-Report Scale-V1.1 (ASRS-V1.1). The BART presented two colors of balloons with differing probabilities of exploding, and participants were incentivized to pump the balloons as many times as possible without causing them to explode. The primary endpoint was "mean adjusted pumps", determined as mean across trials of the number of pumps on trials that did not end in explosion. An index of reactivity to loss was calculated as the difference between the mean adjusted pumps following an explosion and the mean adjusted pumps following trials in which the balloon did not explode.

The ADHD and control groups did not differ on mean adjusted pumps across trials, but they did differ in their reactivity to explosion of balloons that followed the most pumps, incurring the greatest level of loss (F(1, 551) = 7.1, p < 0.01). Interestingly, ADHD participants showed a greater reactivity to loss on these balloons than controls (p < 0.05), indicating that they reduced their number of pumps following balloon explosions more than controls. For participants as a whole, there were small correlations between loss reactivity and scales of everyday impulsivity on the BIS-II (ps < 0.05). For ADHD participants, loss reactivity was unrelated to symptoms of inattention but was significantly correlated with symptoms of hyperactivity/impulsivity (p = 0.01) and total ADHD symptoms (p < 0.05) on the ASRS-V1.1.

In the context of a risky decision-making task, adults with ADHD showed greater reactivity to loss than controls, despite showing comparable patterns of overall performance during the BART. The magnitude of behavioral adjustment following loss was correlated with symptoms of hyperactivity/impulsivity in adults with ADHD, suggesting that loss sensitivity is clinically related to impulsive behavior in everyday life. These findings help to expand our understanding of motivational processing in ADHD and suggest new insight into the ways in which everyday symptoms of ADHD are related to sensitivity to losses and punishments.