|
nif:isString
|
-
Participants were tested in a between-subjects design. On arrival, they were shown either 300 ml or 500 ml of soup. Participants then consumed either 300 ml or 500 ml. An orthogonal combination of seeing either 300 or 500 ml and then eating either 300 ml or 500 ml rendered four separate conditions. ‘Incongruous eating’ was achieved by covertly manipulating soup entering or leaving the bowl during the meal. Appetite was assessed for three hours after the meal and the expected satiation of the soup was assessed approximately 24-hours later.
One hundred volunteers (69 female and 31 male) completed the study and produced responses to an awareness questionnaire indicating that they were unaware that the soup bowl had been modified. Six other participants reported a degree of awareness and were rejected and replaced on this basis. All were staff or students at the University of Bristol. Volunteers had a mean BMI of 23.4 (SD = 3.46), 22 were overweight and five were obese. Participants were recruited by email. Vegetarians and vegans were excluded, together with anyone who declared a food allergy and/or intolerance. All received ten pounds Sterling for their assistance. The study was approved by University of Bristol Faculty of Science Human Research Ethics Committee. All provided written informed consent before assisting with the study.
Soup was added or removed from a transparent soup bowl using a peristaltic pump (see Figure 1). The soup bowl was presented in front of the volunteers and it was fixed to a table. A tall screen was positioned at the back of the table. This separated the participant from both the experimenter and a second table, supporting the pump and a soup reservoir. Throughout the experiment, the volunteers were unable to see beyond the screen.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0050707.g001 Depiction of the self-refilling soup bowl apparatus. The bottom of the soup bowl was connected to a length of temperature-insulated food-grade tubing. This connection was hidden from the participants using a tablecloth. The tubing fed through a hole in the table (immediately under the bowl) and connected to the pump and then to a reservoir of soup via a hole in the screen. The experimenter was able to manipulate the direction and rate of flow using an adjustable motor controller that was attached to the pump. The pre-heated soup was ‘creamed tomato soup’ (supplied by Sainsbury’s Supermarkets Ltd., London; 38 kcal/100 g).
Our measure of expected satiation was based on a ‘method of adjustment,’ and is described in detail elsewhere [27], [28]. Briefly, a 400-ml portion of soup was placed in front of the participant and a photograph of a ‘comparison food’ was displayed on a computer screen. The participants were instructed to match the food picture so it would ‘fill them up as much as the bowl of soup in front of them.’ Participants were instructed to taste one spoonful of soup from the bowl and then adjust the amount of comparison food on the screen. The left arrow-key (on a keyboard) caused the portion size of the comparison food to decrease. The right arrow-key caused the converse. The pictures were loaded with sufficient speed that continuous depression of the left or right arrow key gave the appearance that the change in portion size was ‘animated.’ Each trial started with a different and randomly selected portion size. Participants completed four trials, with each of four comparison foods: chicken tikka masala curry and rice (a dish that is very popular in the UK), margarita pizza, oven-baked fries, and egg penne-pasta mixed with pasta sauce. Comparison foods were presented in a counterbalanced order across participants. Portion sizes of the comparison food images ranged from 50 kcal to 1250 kcal and were spaced in equal logarithmic steps. Fifty images were taken of each food on the same white plate (255 mm diameter), with photographic conditions maintained as constant as possible.
Testing took place individually between 11∶00 and 14∶30 hours. Volunteers attended two sessions approximately 24 hours apart. They were asked to abstain from eating for three hours before the initial session, and to confirm that they had complied with this request on arrival. Hunger and fullness were then assessed using 100-mm visual-analogue scales labelled ‘How [hungry/full] are you right now?’ and anchored ‘not at all [hungry/full]’ to ‘extremely [hungry/full].’ Participants were also asked to report how long it had been since their last meal. Using this assessment of hunger, the participants were then pseudo-randomly assigned to one of the four conditions using a minimisation method [34], [35] with a 4∶1 element of chance. This method made it probable that the groups would be balanced for age, gender, and initial hunger. Participants were then taken to a testing booth where a bowl of soup was waiting. They were instructed to avoid touching the bowl and to eat until the volume of soup remaining matched a line on the side of the bowl. The line ensured that eating terminated with 100 ml of soup remaining, thereby obscuring the bottom of the bowl. To accommodate for this amount, across conditions, the initial starting portion was 100 ml larger than the amount consumed. All participants were informed that eating their prescribed portion was a mandatory part of the procedure. After the meal, hunger and fullness ratings were then taken once again. Participants were then given a pack containing an information sheet with written instructions, a food diary for the rest of the day, and three hunger and fullness rating scales, labelled one-hour, two-hours and three-hours. They were also issued a buzzer that sounded every hour for three hours. On each occasion, they were instructed to complete the appropriate hunger and fullness rating. This procedure has been used in previous studies in our laboratory and compliance with these instructions has been found to be high [36], [37]. Over this three-hour period the volunteers were instructed to abstain from eating and from drinking calorie-containing beverages. Approximately 24 hours later the participants were shown a bowl containing 400 ml of tomato soup and evaluated its expected satiation. They then completed the Dutch Eating Behaviour Questionnaire (DEBQ) [38] which comprises three sub-scales that assess aspects of everyday dietary behaviour (dietary restraint, external eating, and emotional eating). The participants then completed a two-part questionnaire to assess demand awareness. The first section required participants to guess the purpose of the study. In the second section the participants were asked to indicate ‘yes’ or ‘no’ in response to the question ‘was the soup manipulated in any way?’ Participants who selected ‘yes’ were then asked to explain their response. Finally, a measure of their height and weight was taken. Debriefing took place by email, after all of the data had been collected.
One-way ANOVA was used to explore evidence for differences in baseline characteristics across conditions. Specifically, we assessed BMI, age, initial hunger, and scores on the three subsections of the DEBQ. To explore hunger ratings in the inter-meal interval, we used a mixed-model ANOVA with time (0, 60, 120, and 180 minutes) as a within-subjects factor and amount seen (300 ml or 500 ml) and amount eaten (300 ml or 500 ml) as between-subjects factors. The same approach was also used to analyse ratings of fullness. Where we found a significant main effect or interaction term we used ANOVA to scrutinise the effects of perceived and actual amounts, at each time point, separately. In all cases, to reduce error variance, we included baseline ratings as a covariate where a significant correlation existed between a dependent measure and its baseline counterpart. Finally, for each participant, we calculated an expected-satiation score (kcal) by taking an average (mean) of the selected comparison foods. Higher values indicate that the soup was expected to deliver greater satiation in the second test session. To explore the effects of amount seen and amount eaten we submitted these expected satiation scores to a 2 x 2 ANOVA.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}