<p>Although author self-citations are an inherent aspect of scholarly practice, they can be manipulated to inflate some authors’ indices. The objective of this study was to determine the author self-citation rates within the extreme publishing authors (≥ 1 paper/month) in COVID-19 research from 2020 to 2023. This study utilized Scopus data and a further analysis was performed in SciVal. Statistical tests included Mann-Whitney and Kruskal-Wallis tests. Information from 354 unique authors was retrieved. The median self-citation rate from this sample among extreme publishing authors was 3.39%. The highest author self-citation rate was 45%. Men tended to self-cite more often than women (<i>p</i> &lt; 0.001). More self-citations were found on articles rather than reviews or other types of publication (letters, editorials, notes, and short surveys) (<i>p</i> &lt; 0.001). These findings indicate that both publication format and gender shape citation behavior, with original research articles showing higher self-citation incentives. Using the composite citation index, authors who have a lower c-score tend to self-cite more often than reputed authors (<i>p</i> &lt; 0.001), revealing a possible mechanism of the Matthew effect. This pattern suggests that self-citation may be used by less recognized authors to increase their visibility, although it can also reflect legitimate continuity in their research. There was a weak to moderate correlation of author self-citation with other bibliometric indicators. This study contributes to a better understanding of the self-citations of prolific authors in COVID-19 research, while also highlighting potential areas for improvement in the ethical consideration of this topic during the pandemic.</p>