To discuss the reliability of eyewitness testimony in general terms is very misleading. A civil or criminal trial is often long after the event - sometimes years. In the interim, witnesses have been interviewed or deposed, prepped by lawyers, influenced by conversations and media coverage, emotionally bonded with one side or the other. We see this in Mark Lane's interviews and the way that many prominent JFKA witnesses' stories got markedly "better" over the years and decades. Studies with hoax UFOs (i.e., designed to fool witnesses) showed that people accurately described what they saw (e.g., three flashing lights in a triangular formation) but simply misinterpreted it as otherworldly - but, then, the study had been designed to fool them by having the phenomenon not easily fit any known aircraft. With a witness's description right after an event, we are really not talking about "fallibility of memory" but simply "accuracy of perception." With the JFKA eyewitnesses and earwitnesses, we'd have to take into account the sudden, very short and wholly unanticipated nature of the event and the circumstances of each eyewitness or earwitness - location, what he or she was doing at the time, etc., etc. To simply "count heads" is likely to be very misleading. Harold Norman thought the "first shot" sounded different and seemed really only certain about the second and third - and the rifle was supposedly right above his head! Regarding Glen Bennett, one would think his training would make him a solid witness - but he said he heard "what sounded like a firecracker." Given the number of witnesses who said the "first shot" sounded different from the others, I'm not sure that we can reasonably attribute these descriptions to a missed gunshot - especially given the lack of other other evidence for such a gunshot and the wildly different estimates as to when it occurred.