O.K. Marco, that passage you quoted is a fine example. It basically empties the bit bucket for possible causes of reactivation of HERVs. What is definitely observed is that one or more of these factors are present in those cases where virions or other signs of reactivation are also present. Simple biochemical causes can indeed cause transcription of these genes. However, with the possible exception of HERVs of very recent origin, all endogenous sequences are defective w.r.t. to replication. A helper virus may supply missing functions, but we have to ask what possible helper viruses could perform the particular functions required. Optimally, it should be very similar. Retroviruses which have been studied commonly affect immune function to various degrees. This makes activation by other viruses hard to distinguish from a common cause of activation and increased activity by endemic viral infections. A smoke screen of coinfections which hides a retrovirus from immune response will also interfere with detection by researchers. The cases I know of in which a sequence passed via inheritance is easily activated and replication-competent are things like chromosomally-integrated HHV-6, which really does exist. Activation is no problem because exogenous HHV-6 is widespread. The question of how those virus genomes got inserted in chromosomes remains up in the air, but coinfection by a retrovirus is one possible answer. Another possibility involves retrotransposons. Either one is an example of horizontal transmission of genetic information, which has generally been ignored in genetic studies. You can find other examples of chromosomal integration of viruses, though all I know are pathological. We come back to the evolutionary bind, if those sequences are not replication-competent it will be hard for them to evolve. Changes, whether favorable or not, will not be passed on. If it only required a single mutation to produce a replication-competent virus from an endogenous sequence, then we could expect this to happen by chance once in a while, after transcription is activated. Many convergent mutations are another kettle of fish entirely. Arguments about sequences being close enough to be activated and corrected by chance cut both ways. If the correction can take place on a short human timescale, the previous evolutionary history of that sequence must also be short. Either alternative leads to the reasonable idea that undetected human retroviruses are active in the present or very recent past. Your other questions are highly pertinent, and deserve answers I do not have. Your material about sensory gating of visual or auditory information is highly relevant to a possible overlap between pathology of ME and schizophrenia. Again, I suggest that the ability to reject false input and accept only real data from the external world may depend more on the specific parts of the nervous system infected than any fundamental difference in etiology. This is speculation, but it could be at least as productive as a great deal of earlier speculation which achieved academic respectability without doing anything useful for patients in any diagnostic category.