Acute Stimulant Treatment and Reinforcement Increase the Speed of Information Accumulation in Children with ADHD

Abstract

The current studies utilized drift diffusion modeling (DDM) to examine how reinforcement and stimulant medication affect cognitive task performance in children with ADHD. In Study 1, children with (n = 25; 88 % male) and without ADHD (n = 33; 82 % male) completed a 2-choice discrimination task at baseline (100 trials) and again a week later under alternating reinforcement and no-reinforcement contingencies (400 trials total). In Study 2, participants with ADHD (n = 29; 72 % male) completed a double-blind, placebo-controlled trial of 0.3 and 0.6 mg/kg methylphenidate and completed the same task utilized in Study 1 at baseline (100 trials). Children with ADHD accumulated information at a much slower rate than controls, as evidenced by a lower drift rate. Groups were similar in nondecision time and boundary separation. Both reinforcement and stimulant medication markedly improved drift rate in children with ADHD (ds = 0.70 and 0.95 for reinforcement and methylphenidate, respectively); both treatments also reduced boundary separation (ds = 0.70 and 0.39). Reinforcement, which emphasized speeded accuracy, reduced nondecision time (d = 0.37), whereas stimulant medication increased nondecision time (d = 0.38). These studies provide initial evidence that frontline treatments for ADHD primarily impact cognitive performance in youth with ADHD by improving the speed/efficiency of information accumulation. Treatment effects on other DDM parameters may vary between treatments or interact with task parameters (number of trials, task difficulty). DDM, in conjunction with other approaches, may be helpful in clarifying the specific cognitive processes that are disrupted in ADHD, as well as the basic mechanisms that underlie the efficacy of ADHD treatments.

Keywords: ADHD; Diffusion model; Drift rate; Methylphenidate; Reinforcement.

Figure 1.. Diffusion model parameters.

Hypothetical trial depicting the drift diffusion model (reprinted from Moustafa et al., 2015). Drift rate (v) represents the rate of information accumulation, with a steeper slope indicating faster information processing. Boundary separation (a) indicates the level of response caution; wider boundaries signify a more cautious response style. Nondecision time (encoding and motor time, Ter) encompasses time unrelated to decision making, including time for stimulus encoding and response execution.

Figure 2.. Task structure for the baseline and reinforcement version of the XO task.

Task structure for the baseline XO task used in Visit 1 of Study 1 and Study 2 (Fig 2a), and the no-reinforcement (Fig 2b), and reinforcement (Fig 2b) versions of the XO task used in Visit 2 of Study 1.

Figure 3.. Diagnostic group differences and reinforcement and stimulant medication effects on drift rate.

Baseline refers to Visit 1, and both the reinforcement and no-reinforcement drift rates are from Visit 2. The same 100-trial baseline task used in Study 1 was used in Study 2, allowing for baseline (Visit 1) - Study 2 comparisons. The reinforcement manipulation task used in Visit 2 included four 100-trial blocks that alternated between reinforcement and no reinforcement (200 trials per condition).