Bubblbu
commented
3 years ago Scite have faulty data here: "10.1093/brain/awx285"
https://www.e-enm.org/journal/view.php?doi=10.3803/EnM.2017.32.4.413 and https://www.nature.com/articles/nrm.2016.116 seem to have a pretty close relationship based on their 8 citations + very similar abstract.
Some citations are duplicated because of minor changes in the text snippets.
This is clear when the citations are once collected from the direction of seed articles, and once again from the perspective of the citing articles.
Solution could be to ask Scite for the latest set of extracted citations for each article as it seems to me that some citations derive from mutliple rounds of text processing.