At last summer's meetings of the American Sociological Association I ended up at a talk discussing a methodological development for getting accurate data regarding people's movements (how or why I was there I don't recall, and I'm too lazy to dig up last year's program and remember who was doing the presenting). The relevant point is that this group of researchers, who if I remember correctly were interested in the daily routines and activities of young people, were trying to get around a classic problem of social research: people don't necessarily have particularly good recall about what has happened to them. It's simply a function of how memory works; there are scores of psychological and sociological studies on how the brain operates, how memory operates, how our brain often fills in missing details with what we think should have happened, etc.
So to get around this problem, they issued every participant in their study a GPS-enabled smart phone that they were to keep on themselves at all times. To show how much more effective and accurate this method was compared to self-reporting, the presenter showed several examples of the GPS-generated maps of people's movements against their self-reported activities.
It turns out the two were often quite different. I remember one example in particular in which someone's self report simply read that they came home from school, then a couple of hours later went to the museum with a friend, and then went home again. But the GPS map showed they ad actually stopped at a corner store on the way home, rode their bike around aimlessly for awhile, went to their friends house before going to the museum, and a few other little trips around the neighborhood.
The point of these examples were to demonstrate how faulty human memory is about most things we do and experience, even events that happened the previous day (think about it yourself: try to list everything you did yesterday. I'm pretty confident you'll find there are rather sizable chunks of time you can't fully account for). And it wasn't like these study participants were intentionally lying; not only was there no discernible motivation for lying about, say, riding your bike around, but when presented with the discrepancies, most participants immediately remembered the events they had forgotten and even apologized to the researchers for their forgetting.
It's important to remember these omissions happened during normal days in which nothing extraordinary happened to these folks. So what does this have to do with crime and rape culture? Well, I was reading a piece the other day on the on-going, mistake-filled investigation of sexual assault charges against Florida State quarterback Jameis Winston. For those of you not familiar, it appears by all accounts to be a classic case of multiple levels of university and public officials looking the other way when a famous athlete has been charged with rape (as the article linked to points out, at this point the only options are that the University and local PD had either intentionally covered this up/dragged their feet on the investigation OR they are so incompetent as to force the question of why any of them are still employed).
But Winston's defenders, who are quick to flood the comments section of any article on the subject, continually point to the fact that the victim's story has changed some from when it was first reported to the police. Granted, these people are pretty clearly folks who already don't believe Winston's accuser and are working backwards to find socially acceptable reasons as to why (most such comments I've read go on to accuse the woman of being a gold-digger, which is not only patently offensive on it's own, but makes zero sense. In all my years studying the criminal justice system, I've never once heard of someone being ordered to pay financial restitution to a person they are convicted of sexually assaulting).
But as the research above (and a giant line of research before it) makes painfully clear is that human memory doesn't work like that. After all, if people have difficulty remembering what happened to them on a normal day, it's not too big of a stretch to say people who have just experienced a traumatic event in which they were possibly drugged would have an even harder time.
But a rational human being who doesn't start from the position that all women are jezebel whores who deserve whatever happens to them doesn't even need the piles and piles of empirical evidence proving human memory does not work like that. Because again, you can try it yourself: the next time you have a night out at the bar, wake yourself up at 4 a.m. and then immediately state every detail of the previous night. Unless you have a photographic memory, I guarantee you will not be able to account for every minute of the evening. And then I further guarantee that if you were to come back to that account a few days later, you will probably have had time to remember some other things that happened, or forget details of things you previously remembered. Now imagine reporting that story to an officer that openly lets you know they don't believe you and being fully aware that any small mistake you make will result in thousands of shit heads calling you a gold-digging whore online and in real life...