Maximilian Philipp Wolkersdorfer

• Since its conception, experimental psychology has been on a quest to unravel the cognitive processes involved in human behavior. Of major interest in this endeavor is not only the identification of said processes, but to also reveal their temporal dynamics. In other words, can we identify at which point in time which processes are at play, and how they correspond to human behavior? Historically, the fundamental measures when researching the temporal dynamics of human behavior have been, and are to this day, reaction or response time (RT) and accuracy, respectively. In experimental psychology, RT most often refers to the time passed between stimulus presentation and an associated response. Accuracy then evaluates the performance of this action, on a scale from “0 (incorrect)” to “1 (correct)”, depending on the task at hand. Despite the emergence of neuroimaging techniques such as electroencephalography (EEG), magnetoencephalography (MEG), and functional magnetic resonance imaging (fMRI), RT analyses remain a key tool in the field. The aim of this work is to give a brief overview of the historic developments associated with these measures (Chapter 1; based in parts on Article 1 & 2). Thus, I discuss their properties and application, and the methods created to use these measures as a tool to better our understanding of human cognition and behavior. This reaches from Donders’ subtraction method to Sternberg’s additive factor method. Following this, more recent developments are introduced, focusing on the distributional properties of RT and accuracy, such as quantile plots and Vincentizing. Finally, I make my own case for the analyses of the full distribution of response occurrences (Chapter 2; based on Article 1). To do so, I employ discrete-time event history analysis to various experimental psychology tasks (introduced in Chapter 3; see Articles 1–4). I aim to highlight its advantages, the gains and insights it lends to the field, discuss its shortcomings, and suggest potential developments in the future (Discussion; Articles 1-4). Overall, I show that event history analysis (1) can reveal effects that mean performance measures can conceal, (2) can reveal the temporal dynamics of response behavior, and (3) allows to track performance changes on multiple time scales.