I'm reading a great book called "Influence: Science and Practice" by Robert B. Cialdini. It's full of research and anecdotes about how to influence people. It's a real eye-opener.
One of the most potent forms of persuasion has to do with people's innate need to be consistent. Studies show that people will ignore logic and information to be consistent. (In other words, we are moist robots.) According to the research, humans are hardwired for consistency over reason. You already knew that: People don't switch political parties or religions easily. What you didn't know is how quickly and easily a manipulator can lock someone into a position.
For example, researchers asked people to write essays in support of a random point of view they did not hold. Months later, when surveyed, the majority held the opinion they wrote about, regardless of the topic. Once a person commits an opinion to writing even an opinion he does not hold it soon becomes his actual opinion. Not every time, but MOST of the time. The people in these experiments weren't exposed to new information before writing their contrived opinions. All they did was sit down and write an opinion they didn't actually have, and months later it became their actual opinion. The experiment worked whether the volunteers were writing the pro or the con position on the random topic. (Most of the truly stupid things done in this world have to do with this consistency principle.)
I've always been bad at arguing for points I don't believe in — I'm thinking back specifically to a whole-class debate exercise on euthanasia about ten years ago. I wasn't on the side of those arguing against it, but felt very sorry for those who were.
I can see how this would work from the general perspective... reviewing things, be they music, books or whatever, I tend to appreciate them more than if I hadn't. A partial corollary is that that no-one tends to review stuff they don't like or dislike, unless they've already committed themselves to doing so, or they're made to. That element of obligation being there, though, the survey results seem reasonable for a topic such as writing about a person, even a person initially disliked. It's already well-known that touch — such as hands accidentally brushing — tends to leave a more favourable impression, as long as there isn't a strong existing negative association working against.
(Scott Adams scares me, almost as much as rusting metal in black water. Tacked onto the end of a book about his life and drawing Dilbert — The Dilbert Future I think it was — was a pseudo-religious closing section on the subject of affirmations. Not just to do with building up confidence, but affecting the outcome of events.)