How long can an idea stay tantalizing? Back in 2003, I blogged about an experiment that suggested, incredibly enough, that our long-term memories are encoded by prions— the misfolded proteins that are generally accepted to be the cause of mad cow disease. The evidence came from studies of a protein (known as CPEB) that plays a key role in laying down memories in neurons. Scientists found that it had a structure much like prions. When a normal protein misfolds and becomes a prion, it acquires the ability to lock onto other proteins and force them to misfold in the same way. The misfolding can spread until it has devastating results—as in the case of mad cow disease, in which prions from cow brains get into our own brains. But the discovery of prion-like memory proteins hinted that maybe they could play a beneficial role as well. Not long after I blogged on this research, I ran into a neuroscientist I know (and who shall remain nameless). He sneered at the prion paper, pointing out that the authors of the paper didn't show that the protein acts like a prion in neurons. Instead, they had only shown that it acts like a prion when it is inserted into yeast. They took this peculiar step because yeast have prions, and they had the tools to study prion behavior in yeast. It is far harder to experiment with prions in neurons. But this neuroscientist I spoke to thought they shouldn't have gone public until they had taken this last, hard step. I've been waiting ever since. And in the June issue of Nature Review Genetics I came across a paper entitled "Prions as adaptive conduits of memory and inheritance." One of the co-authors is Susan Lindquist of MIT, one of the scientists who made the memory-prion connection back in 2003. Eager for an update, I read on. And what do I find? There's a lot of new research on the role of prions in yeast, where they may play an important role in evolution. But as for prions and memory, there's nothing beyond what Lindquist had to offer in 2003. My patience has probably been irreparably damaged by today's minute-by-minute news cycle, but I have to wonder why we're still in prion-memory limbo. Is the next experiment too hard to do? Does it take years to finish? Or is the link between memories and prions just not there? Just as I'm tempted to give up hope, out comes another paper. It may not seal the deal, but at least keeps me eager for more. Psychiatrists in Switzerland were inspired by the original prion-memory experiments to look for evidence in people's genes. Some studies have suggested that the strength of people's memories is at least partly the result of genetic variation. But no one knew which genes were involved. So the psychiatrists took a look at the prion protein gene (PRNP), which causes mad-cow disease when it misfolds. (No one is sure what it does for us in its normal shape.) People have different versions of PRNP, some of which are more prone to misfolding than others. The scientists genotyped 354 subjects to see which version they carried and then gave them a memory test. In a paper in press at Human Molecular Genetics, they report that people with one or two copies of the misfolding version recalled 17% more information than those without a copy. It's a puzzling result for many reasons, not the least of which is the fact that the link originally proposed between prions and memory did not involve PRNP but CBEP. But it's enough to keep me wanting more.