This tutorial reviews functional connectivity analysis methods and their interpretational pitfalls. It discusses metrics for functional connectivity, including coherence, phase synchronization, phase-slope index, and Granger causality, and highlights common pitfalls such as the common reference problem, signal-to-noise ratio problem, volume conduction problem, common input problem, and sample size bias problem. These pitfalls are illustrated with MATLAB scripts that simulate these issues. The paper also discusses how these issues can be addressed using current methods. It provides a taxonomy of functional connectivity metrics, including model-based and model-free approaches, and discusses various analysis pitfalls that can occur when applying these methods to real data. The paper also discusses the importance of distinguishing between metrics that are computed from the time or frequency domain representation of the signals. It reviews measures of synchronization, including coherence, phase locking value, and phase slope index, and discusses their limitations and interpretations. The paper also discusses the use of Granger causality to quantify directed interactions and its limitations. Finally, it discusses the limitations and common problems of functional connectivity methods, including the impact of non-parametric vs. parametric computation of Granger causality and the differences between bivariate and multivariate spectral decomposition.This tutorial reviews functional connectivity analysis methods and their interpretational pitfalls. It discusses metrics for functional connectivity, including coherence, phase synchronization, phase-slope index, and Granger causality, and highlights common pitfalls such as the common reference problem, signal-to-noise ratio problem, volume conduction problem, common input problem, and sample size bias problem. These pitfalls are illustrated with MATLAB scripts that simulate these issues. The paper also discusses how these issues can be addressed using current methods. It provides a taxonomy of functional connectivity metrics, including model-based and model-free approaches, and discusses various analysis pitfalls that can occur when applying these methods to real data. The paper also discusses the importance of distinguishing between metrics that are computed from the time or frequency domain representation of the signals. It reviews measures of synchronization, including coherence, phase locking value, and phase slope index, and discusses their limitations and interpretations. The paper also discusses the use of Granger causality to quantify directed interactions and its limitations. Finally, it discusses the limitations and common problems of functional connectivity methods, including the impact of non-parametric vs. parametric computation of Granger causality and the differences between bivariate and multivariate spectral decomposition.