Physiological and Eating-behavioral Responses to Viewing Sensory-specific Food Pictures

Watching television is associated with an increased risk of being overweight. On the more modern internet-based digital media food represents a major content category. The vast majority of people in the United States look at screens during the majority of their meals. It, therefore, seems prudent to study the eating behavioral impact of common (food-related) digital media habits.

One particularly popular digital media type is food pictures. While food pictures can elicit hunger, they might also cause satiation. One study showed that rating the taste of 60 (vs. 20) food pictures of a salty (vs. sweet) taste decreased liking for subsequently consumed peanuts (i.e., a salty food). The mechanism behind this finding is an implicit mental simulation of food consumption - necessary to make a hedonic judgment - and a process that spontaneously occurs at the sight of food. Analogous research in mental imagery - explicit mental simulation - similarly found that repeated imagined consumption of a food can decrease subsequent actual intake of that same food.

Cephalic phase responses are conditioned or learned physiological anticipatory responses to food. In the subsequent century, scientific investigation has moved beyond saliva to also include hormones measured in blood. Many cephalic phase responses depend on vagus nerve efferent activation. Pancreatic polypeptide is the most robust and reliable proxy for such vagal activity. Researchers have used a multitude of methods to trigger the cephalic phase responses, including actual food consumption, sham feeding (i.e., orosensory stimulation without absorption), and visual, olfactory, and cognitive cues.

There are some indications that cephalic phase vagal activation might be sensory-specific, i.e., that the vagus nerve is differentially stimulated by sweet and salty stimuli. Pancreatic polypeptide responds differently to (modified) sham feeding solid food with palatable or unpalatable levels of sweet or salty taste while keeping the macronutrient content constant. Specifically, the palatable-sweet conditions had a significantly increased pancreatic polypeptide response compared to the salty conditions.

Contrary to the common belief that sweetness is universally liked, researchers have consistently found marked differences in individuals' taste preferences. Recent research identified FGF21 as a key physiological determinant of sweet taste liking. Fasting FGF21 blood concentrations are lower in participants classified as sweet taste likers compared to dislikers, based on a taste preference questionnaire. FGF21 gene variants were also predictive of habitual candy intake. The agreement between FGF21 blood levels and phenotypical assessment methods of sweet taste liking is unknown. Overall, it seems plausible to speculate that cephalic phase responses manifest sensory-specific differences based on both stimulus and participant.

In the last decade, researchers have begun to study cephalic phase responses to food pictures. None of the studies measured vagal activation via pancreatic polypeptide secretion. While the studies that included a post-stimulus meal challenge replicated established beneficial effects on postprandial glucose metabolism, none of them found explicit evidence for a picture-induced cephalic phase insulin response. However, it must be noted that blood sampling schedules were suboptimal and the studies, therefore, unlikely to capture any effect. The study of cephalic phase ghrelin secretion also yielded inconclusive results. Some research showed a significant increase in total ghrelin levels after exposure to food pictures, but not after non-food pictures. However, other research did not confirm this finding. Ghrelin is of principal importance in hunger regulation, and known to cause food intake in humans. In rats, visual cues increase brain activity and food anticipatory behavior similar to exogenously administered ghrelin. And yet, in another study, participants exposed to food pictures before an ad libitum buffet consumed as many calories as those exposed to non-food pictures.

One important limitation of the aforementioned investigations is that all exposed their participants to 45-50 pictures, depicting a large variety of food and tastes. Some researchers used picture sets of food with different macronutrient profiles, but the food tastes still widely varied within each set. In contrast, prior research using actual food as visual (and olfactory) stimuli exposed participants only to a very limited food selection. Using highly varied food stimuli stands in contrast to the stimulus repetition paradigm, as well as a particular study, which exposed participants to food pictures of a specific taste (i.e., sweet or salty). Furthermore, research has long established that food variety increases energy intake, essentially delaying stimulus habituation. This effect appears to be highly psychological. For example, satiation to one kind of jelly bean can be reset simply by recalling the other kinds they had consumed some minutes ago. An experiment controlling for stimulus taste would elucidate the sensory-specific physiological and behavioral responses to food cues.

Finally, cephalic phase responses may relate to individual tendencies in eating behavior. One study investigating the relationship between the cephalic phase insulin response and dietary restraint, disinhibition, or hunger, as measured by the Three-Factor Eating Questionnaire, did not find any association. However, no study to date has investigated the relationship between external eating tendencies and cephalic phase responses. External eating refers to food consumption in response to environmental cues. One might speculate that, beyond purely neurological influences, endocrine cephalic phase responses mediate such behavior.

The study foresees three participant visits. The three visits mainly differ in the administered visual stimuli. The procedure is described below.

Participants will arrive in the morning around 08:00 after an overnight fast. At 08:15 an investigator will draw an initial baseline blood sample. In the following 15 minutes, an investigator will introduce the image viewing task. Participants will be seated in front of a computer monitor, equipped with a screen-based eye-tracker. An investigator will subsequently configure the eye-tracker and allow participants to acclimatize to the setup. At 08:30, an investigator will draw a second baseline blood sample, and subsequently start the 15 minutes image viewing task. At 08:45, participants will be presented with breakfast choices, shown on the computer monitor. Thus, they will freely choose breakfast type and amount at each visit, which an investigator will prepare and serve at 09:00. Biometric data will be collected during the full image viewing and food choice tasks. While waiting for the breakfast to be prepared, participants will watch a neutral video. After breakfast consumption, participants will rate their post-ingestive sensations.

During the first visit, participants will also undergo basic anthropometric measurements.

During the last visit, participants will also complete the Sweet Liker test, as well as fill out selected eating behavior questionnaires.