This article shows an approach to the study of two fundamental aspects of the prepublication of scientific manuscripts in specialized repositories (arXiv). The first refers to the size of the interaction of "standard papers" in journals appearing in the Web of Science (WoS)-now Clarivate Analytics-and "non-standard papers" (manuscripts appearing in arXiv). Specifically, we analyze the citations found in the WoS to articles in arXiv. The second aspect is how publication in arXiv affects the citation count of authors. The question is whether or not prepublishing in arXiv benefits authors from the point of view of increasing their citations, or rather produces a dispersion, which would diminish the relevance of their publications in evaluation processes. Data have been collected from arXiv, the websites of the journals, Google Scholar, and WoS following a specific ad hoc procedure. The number of citations in journal articles published in WoS to preprints in arXiv is not large. We show that citation counts from regular papers and preprints using different sources (arXiv, the journal's website, WoS) give completely different results. This suggests a rather scattered picture of citations that could distort the citation count of a given article against the author's interest. However, the number of WoS references to arXiv preprints is small, minimizing this potential negative effect. The work of the first, second, and third author was supported by Ministerio de Economía, Industria y Competitividad, Spain, under Research Grant CSO2015- 65594-C2-1R Y 2R (MINECO/FEDER, UE). The work of the fourth author was supported by Ministerio de Economía, Industria y Competitividad, Spain, and FEDER, under Research Grant MTM2016-77054-C2-1-P. The authors would also like to thank the referees for their useful comments and references, which helped them to improve the work, especially in Section 5.