Abstracts: Two much?

I picked this up from Barking up the Wrong Tree:
 When two is too many: Collaborative encoding impairs memory

Humans routinely encode and retrieve experiences in interactive, collaborative contexts. Yet much of what we know about human memory comes from research on individuals working in isolation. Some recent research has examined collaboration during retrieval, but not much is known about how collaboration during encoding affects memory. We examined this issue. Participants created episodes by elaborating on study materials alone or collaboratively, and they later performed a cued-recall task alone, with the study partner, or with a different partner (Experiment 1). Collaborative encoding impaired recall. This counterintuitive outcome was found for both individual and group recall, even when the same partners collaborated across encoding and retrieval. This impairment was significantly reduced, but persisted, when the encoding instructions encouraged free-flowing collaboration (Experiment 2). Thus, the collaborative-encoding deficit is robust in nature and likely occurs because collaborative encoding produces less effective cues for later retrieval. 

This goes against my intuition, as I thought the process of teaching things to other people (being forced to explain it) actually improves information retention. 

In general, the best type of studying (in my experience) is a) having a nice lunch and b) learning the information before the night before.