Transcutaneous Auricular Vagus Nerve Stimulation Does Not Accelerate Fear Extinction: A Randomized, Sham‐Controlled Study

ABSTRACT Transcutaneous auricular vagus nerve stimulation (taVNS) has been tested as a strategy to facilitate fear extinction learning based on the hypothesis that taVNS increases central noradrenergic activity. Four studies out of six found taVNS to enhance extinction learning especially at the beginning of extinction. Facilitatory effects of taVNS were mainly observed in US expectancy, less in fear‐potentiated startle (FPS), and not in the skin conductance response (SCR). Suboptimal stimulation parameters may explain the reported mixed results. Also, variability in selected fear conditioning paradigms and statistical power impedes the comparability between studies. This study sought to further test whether taVNS accelerates fear extinction learning as indexed by US expectancy, FPS, and SCR. Similar to most previous studies, we employed a differential fear conditioning paradigm. The left ear of 79 healthy participants was stimulated with either sham (earlobe) or taVNS (cymba concha) during extinction learning. To maximize the beneficial effects of taVNS, the stimulation of the left cymba concha was administered continuously at the maximum level below the pain threshold. Results of the pre‐registered frequentist and exploratory Bayesian analyses indicate that taVNS did not accelerate extinction learning in any of the outcomes. The null results indicate that taVNS with commonly used stimulation parameters does not reliably optimize fear extinction learning. More research is needed to test if the stimulation protocol determines the efficacy of taVNS in optimizing fear extinction learning. This study tested whether continuous transcutaneous auricular vagus nerve stimulation (taVNS) at maximal intensity below pain accelerates fear extinction learning in a large sample. Our clear null findings add to existing mixed results on taVNS and extinction learning and strongly suggest that taVNS with commonly used stimulation parameters does not reliably optimize fear extinction learning in a differential conditioning protocol.