No one can study love in the laboratory, or even human memory -- OK, we understand the parts of the brain connected with memory, but to me the fact that you can even call up an image from 10 years ago, and, boom, there it is in front of your eyes, in your head, that'll never be fully explained to me.

Ironically, exactly this property of memory - the ability to instantly associate memories - has been very convincingly explained by the science of neural networks back in the early eighties if not earlier.

It turns out that to demonstrate this you only need the simplest of neural setups - a two layer network referred to in the literature as a Binary Associative Memory" or BAM, with one layer representing the input and one for the output.

Almost two decades ago the home PCs of many computer enthousasts were easily fast enough to illustrate this by running simple programs able to associate one digital image with another. BYTE magazine ran an article by Bart Kosko titled "Constructing an Associative Memory" in September of 1987, in which the entire concept was implemented with a mere dozen or so lines of pseudocode.

I have often suspected that it is this unfamiliarity with, and amazement at, the ease with which the mind performs impressive feats of association between memories, our senses, and higher level concepts and ideas, which lies at the root of many unfortunate beliefs, ideas, and philosophies. From "spooky" intuitions and "improbable" everyday coincidences, all the way to the fringe metaphysical "insights" of many otherwise competent scientists.

As an example take Penrose's ideas associating consciousness with quantum phenomena, or on that same subject, Danah Zohar of "The Quantum Self" fame - admittedly pet peeves of mine but I'm sure I wasn't the only physics major who cringed a little when reading their books.

In my mind, as if by some supernatural insight, this interview with Mary Roach instantly triggered associations with both those writers.

First off, let me say that I'm reading and enjoying the book. The first two chapters were quite refreshing after a lot of the stuff I've read on this subject.

However:

"If that University of Virginia study [in which a researcher set a laptop near the ceiling so that if patients claimed to have an out-of-body experience, he could ask them what image was on the screen], if somebody did see an image on that computer on the ceiling, I don't think you would change the mind of any skeptic out there. I think they would come up with a reason why the study is flawed. People are devoted to their convictions."

This is just annoying! Possibly it's true for 5-10% of the people out there, but the fact is that scientists would like proof that it isn't a trick, please. If someone saw an image on that computer, the first thing I would do is check for a way that the person could cheat. If I altered the experiment some and the person could still see it, okay, that comes closer to proof.

A good scientist or researcher will in fact make sure that all other variables are accounted for. If you carefully arrange the experiment so that no cheating (conscious or unconscious) is possible, then what is left, however improbable, is likely the truth.

The example that frequently comes up is the Meteor Crater out west. There was a theory that it was created by, well, a meteor. Scientists were VERY skeptical about that, so they came in and did a bunch of testing. Now there is little scientific doubt that, yes, it was created by a meteor. And that's not a question of scientists "changing their minds," simply one of science proving out an initially outlandish claim. It happens, and scientists are actually thrilled when it does.

Extraordinary claims require extraordinary evidence. I'm eager to see that evidence myself, but I'm not holding my breath until then, lest I find out the hard way!

Thank you. I'm in the 5% mentioned that needs proof of an afterlife.

I grew up with no religious beliefs at all and as I grew older started the search for some kind of answers that life continues on.

Once I became a parent, I needed to believe that if my children died, they wouldn't 'end'. Reading all the horrible stories in the news of senseless death (of children especially) made me even more determined to find a way to believe that they really continued.

I don't want to believe in hellfire or pearly gates. Reincarnation would be a nifty belief system and I thought the premise of 'Defending your Life' was pretty cool.

I'm just glad to see i'm not alone. I heartily apprecate Mary Roach's incredible inquisitive nature and I'm glad she decided to tackle this subject.

My father died last year and I'm pretty convinced that he's hanging out with Carl Sagan somewhere.

Was this article the transcript of a conversation between two adults or two twelve year olds at a slumber party? Unexplained Valentine's candy? "I loved thinking we had a ghost." Mysterious light switches? "Get out! Hmm."

How could Mary Roach write an entire book without realizing that afterlife "research" is utterly devoid of scientific rigor? Exhibit A is the University of Arizona's Gary Schwartz. This crackpot willfully refuses to understand the problem of Subjective Validation, a common phenomenon which completely undercuts his pitiful claims. A quick trip to skeptic.com will explain Subjective Validation to anyone with 21 grams of curiosity.

How can one do even basic research on the fraud of Victorian mediums and toss off this gem about "Douglas Hume" (does she mean Daniel Dunglas Home?): "He was either an amazing magician or some kind of legitimate ghost, I don't know what the heck." How about "lying son-of-a-bitch," Ms. Roach?

This kind of empty-headed agnosticism is irresponsible. Cretins like John Edward, James van Praagh and Allison Dubois are shameless liars who milk the emotions of traumatized people for an easy buck. Instead of being "startled and impressed" by their cheap carnival tricks, Roach should have joined Penn and Teller in gleefully kicking the bullshit out of them.

Salon should have started this interview by demanding that Roach defend her meaningless definition of "skeptic." Here's a hint. When an author says, "I definitely think it's possible to apply critical thinking and be skeptical and at the same time ignore critical thinking and believe in a ghost in your house," it's time for the interviewer to open a dictionary.