Skin Conductance Prestimulus Response Analyses, Artifacts and a Pilot Study

Previous studies have suggested that the human autonomic nervous system responds to stimuli 2–3 seconds before presentation. In these studies randomly chosen photographs with high and low affectivity were presented to participants. Ensemble averaging of skin conductance in the prestimulus epochs showed a differential response between high and low affectivity photographs. In our protocol the problem of idiosyncratic responses to pictorial stimuli was avoided by using audio startle stimuli. Stimulus type was determined just before presentation by a true random generator. Participants heard 20 stimuli per session with a 50% chance of an audio startle as against a silent control. Our dependent variable was the proportions of 3-second epochs prior to audio and control stimuli in which a skin conductance response, that is a minimum in skin conductance followed by a maximum, occurred. We found a significant effect (N ˆ 125, Z score ˆ 3.27, effect size [ES] ˆ 0.0901 6 0.0275, p ˆ 5.4 3 10¡4). Explanations for this result as an artifact were examined and rejected. We show that a significant result from an average-based epoch analysis in this type of experiment is not a necessary requirement to demonstrate significant evidence for a prestimulus response. We also observed post hoc that the prestimulus response effect was correlated with participant lability (r ˆ 0.472, df ˆ 21, p ˆ 0.011).