Introduction

Feelings of hunger, a natural human instinct, are controlled by various physical and psychological functions. But the interaction of these functions to create hunger and the priority given to certain functions over others in reaction to stimuli present an important subject of study. This study benefits multiple fields, including the understanding of eating disorders and the promotion of the food industry (Marcelino et al., 2001). It also comes in handy when faced with the dilemma of whether or not to eat a very appetizing slice of rich chocolate cake even though you’re not hungry. We, therefore, chose to study the psychological force of a visual food stimulus and its effects on the individual’s feelings of hunger. Past studies have supported the supposition that “the sight and a brief taste of a food actually ‘primed’ the appetite, regardless of energy state” (Lambert et al., 1991). The visual stimulus also appears to have a stronger influence on hunger than other types of stimulus, “for example, visual food cues such as the sight of a plate of lasagna might be more powerful than olfactory cues such as the smell of lasagna” (Lambert et al., 1991). These external stimuli seem to take precedence over the actual physical state of being hungry or full. “Cornell et al. (1989) showed that the sight of food enhanced reported desire to eat it, even when subjects were satiated and this was confirmed by the quantity eaten in further ad libitum consumption” (Marcelino et al., 2001). Based on these observations and previous studies, we hypothesize that feelings of hunger can be manipulated as a psychological phenomenon by the introduction of a visual stimulus. Additionally, we are interested in the effects that gender has on the subject’s reaction to the stimulus. In order to examine this subject, our study includes a two-part survey, one of which includes a visual food stimulus in order to provide a comparison between the two situations.

Methods

All participants in the experiment were Dr. McKeachie’s honors psychology 114 students. For the initial survey given, there were twenty participants. Eleven of the subjects were female and nine were males. Eleven of the students had eaten prior to answering the survey and nine had not eaten beforehand. For the second survey given, there were a total of five participants. All participants were from Dr. McKeachie’s honors psychology 114 class as in the initial survey. Of the five students, four were female and one was a male. Two students had not eaten beforehand and three had eaten food prior to coming to class. Due to lack of participants in the second survey, a third survey, identical to the second, was given during a week and a half later during the same time period as the first survey was administered.

A food stimulus (donut holes) was absent in the initial survey and nearby for the second and third. The food was used as an independent variable to test for level of hunger indicated by students when presented with just a survey and with a survey in the presence of the food stimulus. For the initial survey, the students specified their hunger levels at the beginning of the class by rating it on a scale of 1-5 with 1 signifying the most hunger. Students were also asked if they had eaten before class and how hungry they thought they would be at the completion of the class. When presented with a food stimulus, the students answered the same questions as in the first survey where no stimulus was present. But the students also rated their desire to eat the food on the table and their personal like or dislike for the donut holes.

The two surveys were given at the beginning of classes. The first survey without the food stimulus was given during a 1.5 hour lecture at 10:30AM. The second survey was given during a subsequent discussion section at 10:15AM. The third and finals survey was administered during the same 1.5 hour lecture at 10:30AM as the first. All students received the same survey regardless of their age, gender, or whether they had eaten or not. The students were simply told to fill out the survey honesty and quickly based on their current hunger and eating habits prior to the class. When presented with the food stimulus, the students were allowed to view the donuts for a minute before answering the survey questions.

Results

The majority of individuals in the control survey who ate before class reported “slightly” to “no hunger at all” at the end of class. However, there was a sharp increase of hunger in females during the class. No participants reported hunger at the beginning of class, but at the end there were equal amounts of males and females in the "very hungry" category. Males that reported "Barely" as their level of hunger at the beginning of class shifted their hunger value at the commencement of class.

The results of the second survey with the introduced doughnut stimulus had slightly skewed gender results due to a lower number of male subjects. The level of hunger in females did show a dramatic increase with the doughnut stimuli at the beginning of class and stayed at that level of hunger throughout the class. It should also be noted that the majority of participants, in the free answer section of the surveys, which explained their change in level of hunger during the class, cited that they normally get hungrier at that specific time of day.

Conclusion

As assumed, the level of self-reported physical hunger can be manipulated psychologically by the introduction of a stimulus, in this case doughnut holes. Our results showed an increase in hunger primarily at the beginning of class that continued until the end of class with an immediate response to the stimulus. This level of hunger was self-reported and did not take into account other physical symptoms of hunger such as increased salivation. Due to the abundance of women participants, the results were better displayed with the female gender. An interesting finding in our study was the increased desire for the doughnuts in the women participants regardless of their responses of “only slightly liking” the presented stimulus. Thus, female hunger, along with the hunger of the males, was altered by the visual impact of the savory doughnut holes.

Discussion

There were several unforeseen factors in our experimental procedure that made it difficult to garner accurate results during the first attempt of the second part of our experiment. That part included introduction of the subject pool to the food stimulus. We did not account for a cancellation of classes during our scheduled week of surveys and therefore found it difficult to gather all participants together for the second half of the study and the exposure to the food stimulus. The time difference between the first and the second surveys also was a control that we originally expected to have and did not have. Another unforeseen factor was the occurrence of Passover which made it impossible for many people who wanted to eat the doughnuts we presented because they were not kosher for Passover. We were able to remedy these problems however because of our ability to give out the second survey during the following Monday class, thereby eliminating the time difference of the first and second survey as well as lifting the food restrictions placed on some by Passover.

If one would want to continue or expand upon our research there would be a number of ways in which to do so. They could of course gather a larger subject pool to perform the experiment and thereby gather more accurate results. A control group could be instituted in which no exposure at all to a food stimulus would occur, thus allowing the researchers to compare a person without any exposure to food stimulus to those who were not exposed and then exposed at a later time to the food. Gender differences could be measured as well. Even though we took an interest in the issue of gender in summarizing our results, a whole study could be conducted with that one topic at hand. Furthermore, the actual measuring of physiological symptoms of hunger could be useful in determining the actual levels people were at hunger-wise instead of relying on a self report which could at times be biased.

We found several research articles that studied a similar problem to ours, that being if mental manipulation through exposure to a food stimulus could affect physiological hunger in a person. One such study, performed by Saul Shiffman and entitled “Human Models in Craving Research,” was done on the issue of craving in which craving is defined as “a subjective motivational state analogous to emotions rooted in the idea that cues indicate a state of need” (2000). In other words, craving is the emotional feeling of wanting something determined by certain physiological and mental cues. Craving and hunger are similar in that both reflect the need to ingest something. In this study, subjects were asked to measure their level of craving for a number of things including food, drugs, and cigarettes. From these self-reports it was determined that there are two types of craving: background craving, which is a steady state of wanting something throughout the day and requires no environmental stimuli, and episodic craving in which there are occasional bouts of intense craving triggered by environmental stimuli or affective cues. Our experiment dealt with episodic craving as we introduced an environmental stimuli (doughnuts) to people that had not been craving a doughnut beforehand, thereby initiating their craving. Generally, frequent episodic craving tends to be a precursor to relapse, meaning either going off a diet or having a cigarette, as it is often too hard to resist.

Another interesting study was “Self-Reported Measures of Appetite in Relation to Verbal Cues About Many Foods” done by Michael E. Oakes and Carole S. Slotterback. In this study both men and women college students were exposed to a list of written food cues and asked to rate each of the foods according to their nutritional value. Subjects exposed to the food listed reported increases in hunger, desire to eat, and number of foods currently hungry for as opposed to the control participants who did not report any such feelings of reduced fullness. This study thereby concluded that exposure to food cues increased apparent desire of food. The study also took into consideration the gender differences and surprisingly concluded that there were no great effects involving gender on the level of hunger after exposure to food cues. This study used both real foods and cognitive expression as cues. It was found that both cognitive cues and real foods increased reports of general appetite. It was also concluded that external food cues, no matter what they may be, increase the hunger of both dieters and those that are not watching their weight; however, those exposed to the cues, especially women, reported cravings for lower fat foods more often than higher fat ones after the experiment was conducted. This amusing fact reflects greatly the culture of our society in which eating and desire for food is somewhat looked down upon and thereby remedied by reporting a desire for healthier foods.

A final study pertinent to our is “Determinants of Eating Initiation” in which it was determined that interviewing people at times more or less distant from meals is an effective way to obtain various levels of subjective hunger. This experiment was almost identical to our own. Subjects were exposed to a food stimulus, pizza, and then asked to self report on their hunger levels. It was concluded that hunger has no significant effect on the influence of pre exposure to food stimuli thereby supporting our results that we can manipulate hunger levels through food cues. An interesting point in this study was concerning gender differences. It seems that when women are not hungry, they rate their appetite for pizza lower than males however when they are hungry, women tend to rate their appetite just as strong as males do. The study also showed that actually seeing the food, visual pre exposure, had more of an effect on females than males when measuring increase in appetite. Yet after exposure to the food, there is no gender difference in appetite.

The previous study as well as the others examined all supported our hypothesis that we can mentally manipulate people’s hunger through exposure to a food stimulus. It is also interesting to note that eating motivation, although influenced by the internal emotional state of seeing the food, is also contingent upon the level of hunger the person had before exposure to the food. In other words, if someone is completely full there is very little evidence to suggest that putting their favorite food in front of them will make them eat. All in all, our experiment and results seemed to coincide favorable with previous studies done on this topic. More research should be generated in this field in order to one day develop treatments for the craving of both food and drugs as well as in curing problems like obesity and obsessive compulsive eating.

Bibliography

Lambert K.G., Neal T., Noyes J., Parker C., & Worrel P. (1991). Food-related
stimuli increase desire to eat in hungry and satiated human sujects. "Current
Psychology" 10, 297-303.

Marcelino A.S., Adam A.S., Couronne T., Koster E.P., & Sieffermann J.M. (2001).
Internal and external determinants of eating initiations in humans. "Appetite" 36, 9-14.

Oakes, Michael E., & Slotterback, C.S. (2000). Self-Reported Measures of Appetite in Relation
to verbal Cues about Many Foods. “Current Psychology” 19, 137-142.

Shiffman, Saul (2000). Comments on Craving. “Addiction” 95, S171-S175.

The Psychological Effects of a Visual Food Stimulus on Feelings of Hunger

Ashley Erdman

Katherine Torres

Christopher Smith

Samantha Plesser

University of Michigan

April, 2002