AUTHOR=Jean-Richard-dit-Bressel Philip , Clifford Colin W. G. , McNally Gavan P. TITLE=Analyzing Event-Related Transients: Confidence Intervals, Permutation Tests, and Consecutive Thresholds JOURNAL=Frontiers in Molecular Neuroscience VOLUME=13 YEAR=2020 URL=https://www.frontiersin.org/journals/molecular-neuroscience/articles/10.3389/fnmol.2020.00014 DOI=10.3389/fnmol.2020.00014 ISSN=1662-5099 ABSTRACT=

Fiber photometry has enabled neuroscientists to easily measure targeted brain activity patterns in awake, freely behaving animal. A focus of this technique is to identify functionally-relevant changes in activity around particular environmental and/or behavioral events, i.e., event-related activity transients (ERT). A simple and popular approach to identifying ERT is to summarize peri-event signal [e.g., area under the curve (AUC), peak activity, etc.,] and perform standard analyses on this summary statistic. We highlight the various issues with this approach and overview straightforward alternatives: waveform confidence intervals (CIs) and permutation tests. We introduce the rationale behind these approaches, describe the results of Monte Carlo simulations evaluating their effectiveness at controlling Type I and Type II error rates, and offer some recommendations for selecting appropriate analysis strategies for fiber photometry experiments.