Rethinking control conditions in clinical neurofeedback trials

Double-blind, placebo-controlled sham neurofeedback (sNFB) trials are often regarded as the methodological gold standard for evaluating neurofeedback efficacy. However, this paradigm rests on a false equivalence with pharmacological models of intervention. Like genuine neurofeedback (NFB) and unlike inert placebos, sNFB provides structured sensory feedback—including both contingent and non-contingent reinforcement—that engages core learning mechanisms such as operant conditioning. This feature means that sham conditions are not behaviorally or physiologically inert but partially active, undermining their role as true controls. Moreover, the double-blind design is incompatible with a second essential mechanism of neurofeedback: voluntary self-regulation. To bypass this methodological limitation, research practices distance themselves from clinical standards, and in doing so reduce the active effects belonging to NFB. Together, this misalignment leads to systematic underestimation of neurofeedback’s specific effects and to misinterpretation of null findings. Drawing on parallels from psychotherapy and learning-based neuroscience, this paper argues for a reconceptualization of control design in neurofeedback research. Some suggestions for a revised framework are proposed to allow for the separation of specific from non-specific mechanisms without invoking an impossible inert placebo. Adopting this approach can realign the appraisal of neurofeedback efficacy with the principles appropriate to it as an interactive and adaptive intervention.