Transcranial direct current stimulation (tDCS) is becoming an increasingly popular technique for altering eating behaviors. Recent research suggests a possible eating behavior trait-dependent effect of tDCS. However, studies recruit participant populations with heterogeneous trait characteristics, including "healthy" individuals who do not present with eating behavior traits suggesting susceptibility to overconsumption. The present review considers the effects of tDCS across eating-related measures and explores whether a trait-dependent effect is evident across the literature. A literature search identified 28 articles using sham-controlled tDCS to modify eating-related measures. Random effects meta-analyses were performed, with subgroup analyses to identify differences between "healthy" and trait groups. Trivial overall effects (g = À0.12 to 0.09) of active versus sham tDCS were found. Subgroup analyses showed a more consistent effect for trait groups, with small and moderate effect size (g = À1.03 to 0.60), suggesting tDCS is dependent on participants' eating behavior traits. Larger effect sizes were found for those displaying traits associated with study outcomes (e.g., heightened food cravings). "Healthy" individuals appear to be unresponsive to stimulation. Based on this meta data, future work should recruit those with eating behavior trait susceptibilities to overconsumption, focusing on those who present with traits associated with the outcome of interest.

ObjectiveThis study aimed to consider the effect of differing transcranial direct current stimulation (tDCS) parameters on eating-related measures and how issues with experimental design (e.g., inadequate blinding) or parameters variation may drive equivocal effects.MethodsLiterature searches were conducted across MEDLINE, PsycINFO, Scopus, and Science Direct. Studies using conventional sham-controlled tDCS to modify eating-related measures in adult human participants were included. A total of 1135 articles were identified and screened by two independent authors. Study quality was assessed using the Risk of Bias tool. Random-effects meta-analyses were performed, with subgroup analyses to determine differences between parameter sets.ResultsWe identified 28 eligible studies; 7 showed low risk of bias, with the remaining studies showing bias arising from issues implementing or reporting blinding protocols. Large variation in applied parameters was found, including montage, current intensity and density, participant and researcher blinding, and the use of online or offline tasks. The application of differing parameters seemed to alter the effects of tDCS on eating-related measures, particularly for current density (g = −0.25 to 0.31), and when comparing single-session (g = −0.08 to 0.01) versus multisession protocols (g = −0.34 to −0.29). Some parameters result in null effects.ConclusionsThe absence of tDCS-mediated change in eating-related measures may be driven by variation in applied parameters. Consistent application of parameters that seem to be effective for modulating eating behavior is important for identifying the potential impact of tDCS. Using the findings of this review, we propose a series of parameters that researchers should apply in their work.