Attention-deficit hyperactivity disorder (ADHD) is a neurobiological condition that affects both children and adults. Microstate (MS) analyses, a data-driven approach that identifies stable patterns in EEG signals, offer valuable insights into the neurophysiological characteristics of ADHD. This review summarizes findings from 13 studies that applied MS analyses to resting-state and task-based brain activity in individuals with ADHD. Relevant research articles were retrieved from electronic databases, including PubMed, Google Scholar, Web of Science, PsychInfo, and Scopus. The reviewed studies applied MS analyses to explore brain activity differences in ADHD populations. Resting-state studies consistently reported alterations in MS organization, with increased duration (MS-D) and changes in temporal dynamics (MS-C), potentially reflecting executive dysfunctions and delayed maturation of the default mode network. Additionally, MS B demonstrated promise in distinguishing between ADHD subtypes based on differences in visual network function. Task-based and event-related potential (ERP) studies, using paradigms like the continuous performance task (CPT) or Go–NoGo Task, identified MS abnormalities (i.e., N2, P2, P3, CNV) linked to inhibition and attentional resource allocation. Preliminary evidence suggests that MS analyses hold potential for distinguishing individuals with ADHD from control groups. The integration of machine learning techniques holds promise for improving diagnostic accuracy and identifying ADHD subtypes, while MS analyses may also help monitor the effects of stimulant medications like methylphenidate by tracking neurophysiological changes. However, this review highlights the need for more standardized methodologies to enhance the generalizability and replicability of findings. These efforts will ultimately contribute to a deeper understanding of the neurobiological mechanisms that underlie ADHD.