Transcutaneous auricular vagus nerve stimulation (taVNS) is a non-invasive neuromodulation technique that modulates the noradrenergic activity of the locus coeruleus (LC). Yet, there is still uncertainty about the most effective stimulation and reliable outcome parameters. In a double blind, sham-controlled study including a sample of healthy young individuals ( = 29), we compared a shorter (3.4 s) and a longer (30 s) stimulation duration and investigated the effects of taVNS (real vs. sham) on saliva samples (alpha amylase and cortisol concentration), pupil (pupillary light reflex and pupil size at rest) and EEG data (alpha and theta activity at rest, ERPs for No-Go signals), and cognitive tasks (Go/No-Go and Stop Signal Tasks). Salivary alpha amylase concentration was significantly increased in the real as compared to sham stimulation for the 30 s stimulation condition. In the 3.4 s stimulation condition, we found prolonged reaction times and increased error rates in the Go/No-Go task and increased maximum acceleration in the pupillary light reflex. For the other outcomes, no significant differences were found. Our results show that prolonged stimulation increases salivary alpha-amylase, which was expected from the functional properties of the LC. The finding of longer response times to short taVNS stimulation was not expected and cannot be explained by an increase in LC activity. We also discuss the difficulties in assessing pupil size as an expression of taVNS-mediated LC functional changes.