The wavelet transform (WT) is an analytical tool that allows the decomposition of complex physiological signals into their respective spectral components, showing better performance than the fast Fourier transform. WT has been extensively applied to many physiological time series, including perfusion signals, where it has contributed to a deeper understanding of the underlying mechanisms of flowmotion regulation. However, there is a considerable heterogeneity in terms of application of the WT among different authors, which often leads to highly different conclusions regarding the same physiological mechanism. Our objective was to test different approaches regarding the application of WT to skin blood flow signals during a classic maneuver to evoke the venoarteriolar reflex (VAR). In particular, we aimed to clarify whether the characteristics of the original signal influenced the WT output and its interpretation. Fifteen healthy subjects (22.4 ± 5.2 y.o.) participated in this study after giving informed consent. After acclimatization, subjects performed a protocol to evoke VAR on the upper limb while sitting upright – 7 min resting with both arms at heart level (phase I), 5 min with one random arm (i.e., test limb) placed 40 cm below heart level (VAR, phase II) and 7 min recovery in the initial position (phase III). Skin blood flow was assessed in the index finger of the test limb with photoplethysmography (PPG). From the raw PPG signals (PPGr), two new time series were created – (1) PPG amplitude over time (PPGa) and (2) pulse over time (PPGp). All three signals were then processed with the wavelet transform (WT) and decomposed into their respective spectral components. For all WT spectra the dominant frequency and amplitude ratio of each major component were assessed. Both parameters were statistically compared between phases and between signals with the Wilcoxon test for related samples (p<0.05). The PPGr and PPGa spectra showed the same components (cardiac, respiratory, myogenic, endothelial NO-dependent and endothelial NO-independent) in regions with similar dominant frequencies. In contrast, the PPGp spectra only showed components in regions consistent with the cardiac, respiratory and sympathetic activity. The amplitude ratios of the low frequency components (myogenic, sympathetic, endothelial) were significantly different between the PPGr and PPGa spectra during all phases of the protocol. Our results show that although highly valuable as an analytical tool, the WT shows considerable different outputs to different signals, especially in the low frequency components. This suggest that different WT analytical approaches could be considered to extract different information from the same physiological signal.