The psychology of food choice: some often encountered fallacies - Köster (2003) - Article

A new, popular science is the sensory science. This science has a multidisciplinary character and unites psychology, chemistry, food technology and food chemistry to try and understand eating and drinking behaviour as well as trying to make more accepted food products. Every type of science wants to be considered a ‘real’ science and therefore these sciences want to use experimental paradigms. Sensory food science has borrowed experimental paradigms that were valid in other fields, but this was not completely a smart thing to do. In borrowing paradigms from other fields, sensory food science forgot the true nature of the phenomena studied. This was especially the case for the application of psychology. As you all know, psychology is the study of human behaviour and its explicit (conscious) and implicit (non-conscious) determinants.

Sensory psychologists are usually trained in the higher senses (these are visual and audition senses) and they apply the same methods on the lower senses (smell, taste and touch) without really seeing these sensory types as different. Our vision and audition are involved in the spatial orientation and they are strong inborn mechanisms and they guarantee that almost all humans perceive the same direction, forms and distances in a similar way. Our lower senses in the nasal cavity do not have strict inborn mechanisms, but rely on learning. At birth, children don’t like bitter but do like sweet, but later at life we can learn to eat bitter things. This is actually an advantage to our human race, because we are omnivores and can readily adapt to multiple food sources. But there is one consequence and that is that there are large differences in what people perceive and like. Our vision and audition are almost always in the centre of our conscious attention, but the lower senses almost never are. Because of the difference in conscious attention, it is easier to describe visual forms than flavours. But flavours and smells are stronger related to emotions than visual experiences. This may all have serious consequences for the paradigms used in psychological research of the different senses. The current practices in sensory science and consumer research are not always satisfactory from a psychological point of view.

The problems

The problems psychologists are faced with, are that food science research and consumer research are rather diverse. But most of them are related to fallacies that are known to psychologists, because they occur in the same way in psychology.

The fallacy of consumer uniformity

Some people assume that behaviour differs in degree, but not in essence. This fallacy assumes that all subjects in an experiment are comparable and that they perform the same task in the same way. This is one of the most studied fallacies and there is much debate about how to deal with individual differences. Some scientists want new scaling methods, because they may reduce individual variability. There has been not much attention paid to the underlying factors that determine the individual differences.

For instance, one might ask whether people perceive a certain stimulus differently or whether they perceive the task differently. If they perceive the stimulus as different, is this a matter of learning? And if they perceive the task as different, is it due to a difference in understanding the instructions or is it because they use different problem-solving strategies.

When you ask someone how familiar a certain product is, he or she may think that you want to know how often he or she has encountered this product or he or she might think that you want to know how much this product resembles other products.

That person will look into his or her memory, but in the first case he or she will think about the number of encounters with the product, whereas in the other case he or she will scan other products he or she knows for similarity. The question asked should be replaced with a more specific question.

Another mistake based on the uniformity fallacy, is the use of stimulus categories that are based on averaged results from other studies. For instance, in a certain study the writers stated that six pleasant and six unpleasant odours were used. But you need to study the pleasantness of an odour in the group and treat the data on the basis of the individually perceived pleasantness. For instance, one research found that the odour of lavender is liked by the people who recognize it, but disliked by people who don’t recognize it. Also, many studies do not balance for gender and some of them even don’t state the proportion of men and women in the study. Women are more sensitive to odours and flavours than men and also have probably better memory for known odours and this should be taken into account in a study. People should also take into account the different groups. For instance, there are people who bite chocolate and those who suck it. These two groups differ considerably and when one wants to make a certain chocolate product, he or she needs to take these differences into account.

It certainly does pay off the take differences into account. It is however difficult to know beforehand in what respects consumers of a product may differ. This can only be solved by observation and analysis of consumption behaviour and should be tested by analysing a representative group of consumers. Sensory science has come much further with the help of multivariate analysis and segmentation of subjects. But researchers should take into account that situations can influence a person. A certain woman can be seen as a rigid cook when she cooks Sunday dinner for her husband and kids, but she can be seen as a convenience seeker when she prepares food on weekdays after work.

The two situations and their accompanying roles may create very different demands in the same person. Scientists need to ask themselves, how consistent people are.

The fallacy of consumer consistency

The fallacy of consumer consistency is based on the idea that consumers don’t change. Most of the methods in sensory food and consumer science are based on this idea. These sciences want to look decent and assumed that an experiment when repeated under the exact same circumstances will produce the same results. But, what these scientists don’t keep in mind is that consumers really can change. The consumers has a memory and the second encounter with a certain product may not mean the same as the first encounter with that product. In some studies, this won’t matter, but in hedonic judgments and choice processes (in which more personal interest is involved), it will play an important role. It can be shown that changes in preference and choice do take place and in such degree that it can cast a doubt on the predictive validity of hedonic and consumer studies that rely on single measurement sessions. Many recent studies have shown that less than 50% of the participants stayed by their first choice of a certain product.

Psychological theories about choice and preference behaviour also show/predict that there will be a change rather than stability of such behaviour. the mere exposure theory of Zajonc says that the more people are exposed to new stimuli, the more they will like them. There have been studies that found this to be true, but for a number of stimuli there is also a clear decrease of liking to be found. Many researchers therefore argue that Zajonc’s theory represents a special case of Berlyne’s arousal theory. Berlyne made a distinction between diversified exploratory behaviour and exploratory behaviour. In the exploratory behaviour stage a consumer is in a conflictual state because he or she has not much information about the nature of a stimulus and curiosity and novelty will therefore prevail. In the diversified behaviour state, the consumer is no longer in the conflictual state and therefore other stimulus properties, like arousal and stimulus complexity, become predominant. According to Berlyne, there is an inverted U-form relationship between arousal potential (perceived complexity) and liking. This means that for each individual there is an optimal arousal potential level for different stimuli this optimal level is different for different people and depends on learning and experience.

Exposure to stimuli that are a bit more complex than the optimal level would make the subject change his optimal level in the direction the more complex stimulus and exposure to stimuli of a lower than optimal complexity level would leave the optimal level unchanged. Research looking at this has found that products with a stimuli lower than the optimal complexity level leads to product boredom. Product boredom can help explain why certain products flop. According to Walker, prolonged experience with a stimulus, even one of optimal complexity, will reduce its perceived complexity to some extent. If one accepts the idea that this is the cause of product boredom than one should not introduce the most liked product for introduction in the market.

It is better to introduce a slightly less liked, but more complex product. There is also another cause of flops and this is slowly rising aversion. This slowly arising aversion occurs in products that are well liked, but at the same time have a minor defect that raises a slight feeling of doubt in the consumer. The feelings of doubt will grow and the positive feelings about the products will slowly diminish as a result of habituation. Over time, the consumer will dislike the product. This phenomenon is different than product boredom. Product boredom leads to indifference with regard to the product and slowly rising aversion leads to real dislikes. Raising perceived complexity will solve the problem of product boredom, whereas with slowly rising aversion, the source of the minor dissatisfaction must be identified and eliminated.

Single hedonic measurements are good to measure immediate acceptance, but they will not provide information about long term appreciation of a product. There should be combination of repeated laboratory testing with intermediate well controlled exposure will give us that information. Shifts in perceptual measurements do also occur. Certain foods are too complex in flavour to grasp all the aspects of them at once. Sensory adaptation plays a big role in shifts. Also, foods and drinks are eaten in different combinations and this may influence the perception of them.

The fallacy of conscious choice

This fallacy is based on the implicit idea that people are reasonable and make rational choices. Most people see themselves as autonomous and they think that they make decisions based on rational grounds, except in extremely emotional situations. We rationalize our motives and often over-stress certain points in our story to make other people understand us better and we under-stress some socially undesirable aspects of our decisions. In science too, we favour certain hypotheses over others without knowing exactly why. This means that it even may occur that the right hypothesis is never tested or it will get tested, but may have difficulty to get generally accepted or published. In psychology and sensory science, the fallacy of conscious choice becomes dangerous, because putting a question to a subject will always result in an answer, even if the question is actually unanswerable. People may be polite and might not want to disappoint the researcher or they don’t want to look stupid and they think up an answer. Subjects try to guess what the subject of an experiment is and they respond in the way they think the investigator would want to hear. They give socially desirable answers or think of answers instead of saying that they do not know. Because many people give answers, it strengthens the investigator’s conviction that his question was perfectly legitimate. Researchers should not base conclusions on these answers. People usually don’t analyse what they eat or drink, and if they try to do so, they will have difficulty finding words for their feelings and thoughts about a product.

Direct questions, or ‘why’ questions are bad. Indirect questions about certain food preferences are much better, especially when they are about the frequency of behaviour and not about feelings. Direct questions are not always as easy as they seem and can lead to misleading conclusions. This is especially the case when generally accepted ideas about desirable behaviour and about the relation between eating and health are involved. There are certain investigators who report the results obtained from questionnaires about attitudes and values without checking whether these results are related to actual food choice.

Perceptual fallacy

The perceptual fallacy is based on the assumption that what can be observed will be remembered and that what is not perceived can’t be remembered. Most of sensory and consumer research is based exclusively on perception and the role of memory is often forgotten. Even in most studies of experience, one would expect that memory of previous experiences with the product or products that look like that particular product plays a major role, but many researchers remain with only studying perception. They may study the perception of the packaging on the perception of the product, but not what is remembered about the product itself. Research has shown that stimuli that remained implicit and unnoticed in the learning phase, were later remembered. Studies have also shown that unconsciously learned odours and flavours are remembered in everyday life. Everyday experiences show that we have learned more about a product without any previous analysis. For instance, if your favourite chocolate bar tastes different, because the makers have changed certain properties of it, you will realize that you have learned unconsciously a lot about the properties of that chocolate bar, without having analysed all the ingredients and the tastes of all the ingredients. Also, research suggests that what people remember and the way in which they remember it, may depend on the meaning it has to them.

The situational fallacy

The situational fallacy is based on the assumption that perceptual situations are only defined by objective criteria and not by the subjects’ conscious and subconscious intentions. People can change with their intention the situation and the things around you can show different meanings which are closely related to their previous experience with them or their knowledge about them. Things are never without meaning for the perceiver and for psychologists, there are no pure sensations that exist independent of meaningful perception. When researchers ask ‘why’ questions, they will maybe not receive an honest answer (discussed above). True meaning is often based on very intricate and often hidden motives and they usually can’t be brought to consciousness. Also, if they can be brought to consciousness, it is difficult to express them verbally. So meaning can’t be measured by blunt questions, but with indirect and less verbal methods. Hedonic responses need to be measured in a more meaningful context.

There are three misunderstandings, which often occur in research about the influence of context on the perception of food:

Food context: this is the combination with other foods, food packaging, naming and culinary tradition.

The eating situation: the physical and social environment in which the food is eaten.

The individual: food aversions and food preferences, food expectations and variety seeking tendencies.

Researchers can’t ask people directly about these food properties. A situation-oriented approach should be done to figure certain things about food preferences out. In this approach a consumer is not seen as a fixed individual with constant characteristics, but as someone who has many roles in his or her life and who has different intensions and wishes in these roles. In different situations, different roles become apparent. A family dinner is a different situation than dining in a restaurant with friends. Certain cravings, like wanting to eat a chocolate ice-cream, are acceptable in some situations and not in others. For instance, when you’re alone at home you can have that ice-cream, but when you’re at a friend’s house, you can’t just ask for an ice-cream.

But how can situational-oriented research be carried out? Researchers can evoke situations with the help of visual and/or auditory means and give the consumer food. Then the consumer is asked to rate how well this food would be appreciated in the situation and how often he or she would like to eat it in that situation. But there are some difficulties. People may have much more in common in the way they experience situations than it seems at first site. Much of our personal history is influenced strongly by cultural traditions, by the school system and common values and beliefs of our generation.

Communication of feelings would be impossible if situations were really private and completely individual. Another problem is evoking the situations in the laboratory. When you try to evoke situations with visual cues, you need to make sure that all your participants can relate to these cues. People from different cultures, have different eating habits. Some cultures eat with their hands, while others only use forks. Some cultures sit on the floor, while others sit at the dining table on chairs. It is therefore difficult to show visual cues, unless you show general situations. Auditory stimulation is better, because one can imagine his or her own private situations.

The conclusion of the writers is that we need to try to get aware of certain fallacies. There may be a strong dissociation between what we think and do on the one hand and what we actually are and actually do on the other. Most of our actions are implicit and when we think about them, it is difficult to find words for them.