Over the past few days, I’ve been reading Robert A. Heinlein’s The Cat Who Walks Through Walls. In my opinion, it’s a somewhat overrated book, but I’ll admit that it makes for adequate bus reading. The plot of the book I won’t give away, but I will mention that the main characters of the book are frequently eating waffles. Now, I maintain that the waffle is a fine food — its texture (slightly cirspy outside, warm and fluffy inside) and flavor (the slight induction of the Maillard reaction giving it that wonderful brown color and sweet aroma) make for a pleasant eating experience. However, I will not pretend that the waffle has ever been one of my favorite foods.

This said, I have had a rather strong craving for waffles since starting the novel. Were it to persist, I might have been tempted to claim a stress-induced (albeit mild) pica, were it not for the immediately obvious source attribution. What makes this interesting to me as a cognitive scientist, however, is that I have moved from merely craving waffles to both actively seeking waffles (even to the point of encouraging my friends to make them for me) and experiencing phantom waffle sensations, including, but not limited to, believing that I smell waffles (when there are none being made) and having the sensation of tasting recently ingested waffles (which I have not yet done as of the time of writing this). Though I’m sure there is already a name for such phenomena, I suggest the whole group of sensations be referred to as “Phantom Lunch” for conveneince. What makes Phantom Lunch of interest is that it bears some similarity to other phantom sensations. Phantom Limb, for example, is a phenomenon that may occur when a person has been born without a limb or has more recently lost a limb. Well studied by V. S. Ramachandran, sufferers typically report sensation (pain, itching, tingling, etc.) in the missing limb (or within the imaginary bounds that the limb should occupy, if it were intact). In many cases, the sensation is tied to imagining the missing limb in action, suggesting that the areas of the brain responsible for the haptics and kinetics of the limb (particularly the motor and somatosensory cortices) may be interacting with the mental representation of the limb. In support of this, one treatment (pioneered by Ramachandran) for alleviating discomfort in the imaginary limb is the mirror box, which is little more than a box with a mirror that divides it into two compartments. The box is used as follows: let’s say I am an amputee who is suffering from a phantom fist, which is painfully clenched and which I cannot seem to unclench. By clenching my remaining hand and placing that arm into one side of the box, and then placing my other arm into the other side of the box, I see (via the mirror) what appears to be two full arms. Unclenching my remaining hand thus makes it appear as if both hands are unclenching, which has the therapeutic effect of reducing or even eliminating the discomfort I feel in the amputated hand. Essentially, seeing the hand unclench has forced reconceptualization of the representation of the hand as non-clenched. (More in depth information here (pdf).) What makes Phantom Lunch interesting is that phantom food sensations began with recent activation of the concept of the foodstuff in question, and have been intensified by reading additional passages including said foodstuff. This much might be predicted from a basic neural network in which concept and sensory experience are linked. However, what separates it from that scenario is that Phantom Lunch will be indubitably relieved upon actually consuming a physical version of the phantom food. That consummating act may force reconceptualization of the waffle as eaten, and I can go on to live my life without smelling waffles everywhere.

Having not yet sought out this brunch therapy, however, should I be convinced that treatment via viand will be as satisfactory in producing positive results as the mirror box is in more clinical settings? I am inclined to say, “Yes; yes it will be.” My reasons for maintaining this opinion are twofold. In the first, there is the idea that satiation or fullness after consuming the target item will push the system to accept that waffles have been attained and are no longer to be sought out for survival. However, no amount of fullness so far has relieved the craving, so this appears an unlikely mechanism. In the second — our senses are notoriously fickle and prone to adjustment via other-sensory input. Thus it is not unreasonable to think that the phantom smells will be alleviated by tasting waffles, and so forth. As a prime example of this, consider the McGurk-MacDonald effect. Here, the visual modality directly interferes with the aural modality (or vice versa). (If you’re not familiar with the effect, just check out the link.) Slightly less dramatic examples play into taste illusions, where expectations about what something tastes like affects the way people report tasting it. This is of particular importance to the wine tasting industry, which, every now and then, gets punk’d by some meddling psychologists who take joy in getting all up in their hoity-toity bidniz. Other examples abound, even down to the very simple level of eating while your sinuses are stuffed up leading to food tasting different due to disruption of the sensory integration you are otherwise used to. What all this suggests is that the waffle areas of my brain (presumably not too far from the amygdala almonds, the hippocampal bananas, the pineal pine nut…), which are currently overstimulated, might be able to be recalibrated to normal levels via additional input from various sensory sources. In pursuit of this, I could try eating a pad of butter drenched in maple syrup in order to simulate the waffle experience. It would be a bit like virtual reality therapy for phantom limb… albeit tastier, but I somehow don’t expect that to satisfy.

In the end, psycho-economics will probably be the biggest hand at play. Having waffles will eventually saturate my waffle receptors, lowering the utility for future waffle-eating. Eventually, this will reach a point where utility is negative. (For similar discussion of utility see the prior post on cryonic preservation). This should drive my cravings down to a manageable level, where they fall into competition with all my other desires while not rising above significance. I look forward to that, as the desire to consume nothing but waffles is somewhat annoying. I feel like my body should know better. After all, doesn’t the Bible say, “Man cannot live on waffles alone”?

Update:
My usual D&D group gathered today for a brunch session, and we made waffles.  This recipe, specifically (which was quite tasty, I must say!).  In usual fashion, I ate too much and felt slightly ill from having stuffed myself with delicious waffles.  Naturally, this pushes its way into taste aversion territory (though not so strongly as to create a lasting aversion), and all desire for waffles as well as the phantom sensations have disappeared.  This, I feel, must be a particularly common restoration of normalcy regarding Phantom Lunch, as I have found it to be particularly common for people to comment after stuffing themselves on a given food, “Oh, my goodness.  I ate so many/much _____, I don’t even want to think about it.”  If we stuck with the basic neural network description, avoiding thinking about the foodstuff in question would allow decay of the previously activated units, which thus would reduce the phantom perceptions.  The question seems to be, however, what is the driving mechanism here — is it that utility into the negative for additional waffles creates a nulling effect on waffle desirability, thus causing the sufferer to think less about waffles, or is it that discomfort creates a temporary aversion, pushing the system to actively eliminate the stimuli?  Perhaps both or either, depending on specifics of the situation?