Ulric Neisser is probably the person to start with this type of false/malleable memory problems. I have only found interesting things on google book search which do not lend themselves to quoting. Also I couldn't find much more recent studies in this field.