Measuring food reward and the transfer effect of sensory specific satiety

Abstract

The main objectives of our study were (1) to compare several direct and indirect measures of liking and wanting for food and thereby (2) investigating the transfer effect of sensory specific satiety (SSS) for sweet and savory taste to other foods. We used a cross-over design whereby 61 healthy, unrestrained subjects (19M/42F), with a mean age of 21.9 (SD 3.1) y and a mean BMI of 21.7 (SD 1.5) kg/m² were offered a standardized amount of rice meal with either a sweet or savory taste. Afterwards, liking and wanting for 16 snack products, varying in taste (sweet/savory) and fat (high/low), were assessed. Method 1 assessed ad libitum intake, method 2 the willingness to work for access, and method 3 explicit and implicit responses to photographic food stimuli. All the methods used showed a similar pattern of results; after eating a preload with a certain taste, the liking and wanting of snacks with a similar taste were less than for snacks with a dissimilar taste. This transfer effect was not equipotent for sweet and savory tastes. It appears that in young, healthy adults, savory taste has a stronger modulating effect on subsequent food choice than sweet.

Introduction

Sensory properties play an important role in food selection and intake (Blundell et al., 1987, Sørensen et al., 2003). When a food is eaten to satiety, the pleasantness of the sensory properties of that food decreases more than of foods that have not been eaten. This is sensory specific satiety (SSS) and was first demonstrated in humans by Rolls, Rolls, Rowe, and Sweeney (1981).

Not only eaten foods, but also foods that share sensory characteristics of the eaten foods decline in pleasantness relative to foods that do not share these properties (e.g. Rolls, Vanduijvenvoorde, & Rolls, 1984). This has implications for the operation of SSS in a natural setting where availability of foods is unconstrained and varied. In addition, the strength of this transfer effect for different tastes is far from clear. For example, an important distinction regarding taste can be made between sweet and savory, which includes almost 90% of the food we eat (Mattes, 1985). However, the profiles of appetite for something sweet and appetite for something savory show different patterns over the course of a day. It appears that appetite for something savory is more in line with the pattern of meals (more hunger/meal related), whereas appetite for something sweet is more stable during the day (de Graaf, Jas, Van der Kooy, & Leenen, 1993). It is unclear why this occurs.

It has been suggested that SSS is not only represented by a decrease in pleasantness (e.g. ‘liking’: hedonic sensation), but also in ‘wanting’, which refers to the motivation to engage in eating (Berridge, 1996, Mela, 2001). In addition, it is believed that both measures of liking and wanting reflect ‘core’ processes that can operate without conscious awareness (for further reading, e.g. Berridge, 2004, Berridge, 2009, Berridge and Robinson, 2003, Finlayson et al., 2007a, Finlayson et al., 2007b). It is important that we know how these different components of food reward, operating at explicit (conscious) and implicit (unconscious) levels, relate to behavior, to fully understand processes involved in food selection and intake (Finlayson, King, & Blundell, 2008).

To measure these processes of food reward separately, however, is challenging, as many manipulations alter these processes together (Berridge, 1996). There are several direct measurements, e.g. subjective ratings (e.g. Brunstrom and Mitchell, 2006, Weijzen et al., 2008), which typically tap into conscious attitudes. But most often people do not analyze their attitudes towards stimuli consciously. Rather, their behavior is guided by a spontaneous interaction with the environment. Implicit measures are proposed to measure this spontaneous (automatic) reaction to a stimulus. Hence, indirect, implicit measures could be particularly suited to predict spontaneous, uncontrolled behavior (De Houwer, 2009, De Houwer and Moors, 2010).

The definition of an implicit measure is that “the outcome of a measurement procedure is causally produced by psychological attributes in an automatic manner” (De Houwer & Moors, 2010). Many of the implicit measures that have been introduced during the past 20 years are based on reaction time tasks (De Houwer & Moors, 2010). Recently a novel computer-based procedure has been developed, the Leeds Food Preference Questionnaire (LFPQ). Aside from subjective ratings of photographic food stimuli it includes a “forced choice” behavioral measure, whereby the speed with which one stimulus is chosen in preference to its alternative is the indirect measure, proposed to assess implicit wanting (Finlayson et al., 2008, Finlayson et al., 2007a). Another procedure which has been suggested to measure implicit wanting processes is the assessment of a subject's willingness to work (i.e. instrumental responses) to get access to a food (Epstein et al., 2003b, Havermans et al., 2009, Lemmens et al., 2009, Mela, 2006).

In food research actual intake has always been the ultimate measure of (sensory) satiety, and recently it has been proposed that this is also a measurement of implicit wanting processes (Berridge, 2009). The outcomes of above mentioned measures might be tapping more into unconscious processes which are involved in food intake.¹

The main objectives of our study were (1) to compare several direct and indirect measures of liking and wanting for food and thereby (2) investigating the transfer effect of SSS for sweet and savory taste to other foods. We assessed this by measuring, after eating a sweet or savory preload, the ad libitum intake (method 1), the willingness to work (i.e. instrumental responding) for access (method 2), and explicit and implicit responses to photographic food stimuli using the LFPQ (method 3), of several snack products which varied in taste and fat content. By comparing the outcomes of these measures we advance the understanding of the relationship between hedonic and motivational aspects of eating.