When scheduling talks in a conference session, what is the etiquette, if any, regarding placement of multiple talks from the same research group?

You have to plan the symposium from the audience’s perspective, not the individual research group’s.

Presenting multiple talks from the same group back-to-back or even within the same session can become tedious, even if the work is nominally “distinct.” If I were planning multiple sessions, I would avoid putting abstracts from the same group in the same session, but instead spread the pairs out evenly among the three sessions (if possible).

If your sessions are thematic, I’d still try to separate the talks within the session, so that the talks are not back-to-back, but perhaps “bookending” the session or nearly so.

I second aeismail's general advice:

You have to plan the symposium from the audience’s perspective, not the individual research group’s.

However, I derive a somewhat different conclusion from it.

First, order/group the presentations based upon their content, irrespective of who contributed them or who will present, e.g.:

• Order talks in such a way that concepts are covered first, more concrete examples later.
• Group talks by the presented area of application.
• Possibly (if it is well accepted in the field that difficulties of works vary and more complex does not necessarily mean "better") order talks by increasing complexity.
• etc.

Only then, if there is still some question about ordering or grouping, you may want to look at who are the contributors¹. In that case, I'd prefer grouping works whose authors are connected, as that might allow to

• shorten parts of the introduction (the personal introduction if it's the same author, the introduction of the context of application for talks related to the same project)
• lessen the burden on the audience to wrap their minds around totally different contexts of application (all of which may be alien to the audience, despite being familiar with the discussed conceptual questions on a more abstract level)
• get used to one style of presentation (at least if it's the same presenter)
• reduce some confusion among listeners who may try to associate the talks with some external factors in their minds and might then wonder why they remember only (w.l.o.g.) four groups/presenters/projects for five talks. Seeing the two talks by the same presenter in direct succession makes it more obvious what is going on and may thus render it easier to form a mental image of the session structure.

---

¹: I am having a hard time picturing a scenario where being from the same research group (in the sense of a research group as a team working together at one (w.l.o.g.) university with one professor as the lead) could be relevant in any way. Two contributions by either the same presenter (a very special case of "from the same research group") or from the same project/grant (which can, however, involve entirely different research groups from different institutions and parts of the world) seem like a more relevant case in the context of this question.