Human Memory: A Constructivist View (Mary Howes & Geoffrey O'Shea)

The authors present a thorough, tightly-argued case for the idea that memories are constructed, not simply stored. That is, each time we remember something, we use the bits and pieces of information available in the hugely complex network we've built, incorporating traces of all our experiences and knowledge, to build that memory. We don't just go on a shelf and find it there; it's new each time, and is changed every time we remember it. This is a very strong idea and has a huge amount of empirical support. For me, this was the weakness of this book; it argued against the idea that memories are exact replicas of our experiences, perhaps stronger or weaker, faded or intact, but otherwise unchanged. I don't know of anyone in the field who seriously takes that position any more. This made me lose patience with the book, because it seemed they were devoting their energies toward knocking down a straw man rather than actually examining the implications of the constructivist nature of memory. I also had another issue, which is much more mine than theirs; the book leaned heavily toward a philosophical approach, focusing on logic and introspection, but not so much empirical research. It's a matter of taste, I suppose, but that's not to my taste. I can't fault their conclusions, but don't feel that reading it really changed or enlarged my understanding of memory.