ARTICLE IN PRESS Appetite ■■ (2014) ■■–■■

Contents lists available at ScienceDirect

Appetite j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a p p e t

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Research report

Labeling exercise fat-burning increases post-exercise food consumption in self-imposed exercisers Q1 Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer * Department of Sport & Health Management, Technische Universität München, Uptown Munich Campus D, Georg-Brauchle-Ring 60/62, 80992 Munich, Germany

A R T I C L E

I N F O

Article history: Received 27 February 2014 Received in revised form 14 May 2014 Accepted 26 May 2014 Available online Keywords: Physical activity Exercise Food intake Labeling Self-imposed goals Motivation

A B S T R A C T

The goal of the study was to determine whether the label given to an exercise bout affects immediate post-exercise food intake. The authors hypothesized that explicitly labeling an exercise bout ‘fatburning’ (vs. labeling an exercise bout ‘endurance’ exercise) would increase post-exercise food intake in individuals who self-impose physical activity, because they are more likely to see the label as signal of activated fat metabolism and license to reward oneself. No such effect was expected for individuals who do not self-impose physical activity but consider exercise enjoyable. Ninety-six participants took part in an experiment manipulating the label given to an exercise bout (fat-burning exercise or endurance exercise) between participants. They cycled on an ergometer for 20 minutes at a consistent work rate (55– 65% of predicted VO2 max) and were offered ad libitum food (i.e., pretzel pieces) after the exercise bout. The results showed that self-imposed exercisers, that is, individuals with low behavioral regulation and individuals with high psychological distress, high fatigue levels, and low positive well-being when exercising, ate more food after exercise when the bout was labeled fat-burning exercise rather than endurance exercise. The results help develop health interventions, indicating that the tendency to compensate for energy expended following physical activity depends on both the label given to the exercise bout and the degree to which individuals self-impose physical activity. © 2014 Published by Elsevier Ltd.

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Introduction Exercise, in addition to a hypocaloric diet, is the key to weight loss and long-term weight maintenance in humans (Serdula et al., 1999). Individuals who exercise expend energy and thus can reach a negative energy balance more easily than individuals who do not exercise. Individuals may compensate, or even overcompensate, for the energy spent during physical activity via post-exercise energy intake, but the mechanisms for this remain largely unknown (Schubert, Desbrow, Sabapathy, & Leveritt, 2013). King, Tremblay, and Blundell (1997) demonstrated that physical activity generates a drive to eat. In contrast Ueda et al. (2009) found an exerciseinduced suppression of energy intake. Whilst physiological factors and the type and mode of exercise influence post-exercise energy intake (Blundell & King, 1999), there is recent evidence that contextual and psychological factors are relevant too (Hall et al., 2012). This study therefore used a behavioral regulation approach to study contextual effects on post-exercise food intake. The study considers the label given to the exercise bout as one contextual factor that potentially influences post-exercise energy

64 65 66 67

* Corresponding author. E-mail address: [email protected] (J. Koenigstorfer).

compensation. In particular, we consider the influence of labeling an exercise bout ‘fat-burning’ (compared with the label ‘endurance’ exercise) on immediate post-exercise food intake. The label fat-burning may act as signal that fat metabolism has been activated, thus liberating individuals to consume more food after they have finished exercising. This effect is likely to be present in individuals who struggle to regulate their physical activity, because they tend to rely on such signals.

68 69 70 71 72 73 74 75 76

Fat-burning exercise The names of products and services (including brand names) trigger automatic associations in consumers and influence goaldirected behavior (Fitzsimons, Chartrand, & Fitzsimons, 2008). This also applies to food; the names of foods have been shown to influence food intake (Koenigstorfer, Groeppel-Klein, Kettenbaum, & Klicker, 2013; Wansink & Chandon, 2006). For example, labeling trail mix ‘fitness food’ can increase consumption (Koenigstorfer et al., 2013). More pertinent to this study, adding the nutrient claim ‘lowfat’ to foods can cause overconsumption by leading consumers to make erroneous inferences about serving size and reducing consumption guilt (Wansink & Chandon, 2006). The nutrient claim provides a cue to consumers that the food is a low-calorie choice and thus safe to eat. The cue therefore liberates consumers’ attempts to

http://dx.doi.org/10.1016/j.appet.2014.05.030 0195-6663/© 2014 Published by Elsevier Ltd.

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

77 78 79 80 81 82 83 84 85 86 87 88 89 90 91

ARTICLE IN PRESS N. Fenzl et al./Appetite ■■ (2014) ■■–■■

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

control their food consumption (Fishbach, Dhar, & Zhang, 2006; Geyskens, Pandelaere, Dewitte, & Warlop, 2007). References to fat cannot only be found on food packages, but also on products and services in the exercise domain. For example, treadmills and bicycle ergometers offer ‘fat-burning programs’ which individuals can select when they start exercising. Health and fitness clubs offer ‘fat-burning classes’ to their customers. These programs and classes are typically used to describe low-to-moderate intensity exercise bouts. There are reasons to believe that individuals perceive fat-burning exercise to be conducive to positive health and body appearance, including weight loss (Warburton, Nicol, & Bredin, 2006). First, individuals may believe that fat metabolism is stimulated, because the percentage of energy derived from burning fat (rather than carbohydrates or protein) is higher for less intense physical activity. An increase in fat metabolism is associated with several health benefits (Eriksson, Taimela, & Koivisto, 1997). Second, the concept of fat is closely linked to changes in energy balance (Wansink & Chandon, 2006). The fact that individuals burn off fat (and hence calories) may thus be more salient when they exercise using a fat-burning program. These processes likely influence food consumption, given that fat is also often used as an indicator of the calorie content of foods (Wansink & Chandon, 2006). Next, we will explain when we expect fat-burning labeling of exercise bouts to influence post-exercise food consumption.

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Consequences for post-exercise food intake Based on our arguments above, we predict that labeling an exercise bout fat-burning will increase immediate post-exercise food consumption, because it makes the concept of fat burning salient and individuals believe that activated fat metabolism produces health benefits. However, there are reasons to assume that this labeling effect is influenced by the degree to which individuals self-impose physical activity in order to reach health and body appearance goals. In particular, labeling exercise bouts fat-burning may increase food intake in individuals who exercise as a consequence of externally imposed self-regulation, because physical activity legitimizes rewarding activities from their perspective (Markland & Tobin, 2004). In the context of this study, these individuals may perceive exercise in the fat-burning zone as a license to consume and are thus more likely to compensate for the energy expended during exercise. We capture the degree to which individuals self-impose physical activity via two variables: behavioral regulation (i.e., motivational factors) and subjective exercise experience (i.e., emotional factors) (Markland & Tobin, 2004; McAuley & Courneya, 1994). The concept of behavioral regulation implies that individuals pursue an activity as a result of various forms of motivation (Ryan & Deci, 2000). In the context of exercising, behavioral regulation consists of five dimensions: amotivation, external regulation, introjection, identification, and intrinsic regulation (Markland & Tobin, 2004). Amotivation and external regulation are factors that relate to the influence of other people and external rewards; high scores indicate poor behavioral regulation. Introjection, identification, and intrinsic regulation are factors that relate to the internalization of external controls, acceptance of activity as part of the value system, and perception of exercise as enjoyable, respectively; high scores indicate good behavioral regulation. In individuals with high behavioral regulation scores (exercise) behavior is self-determined rather than guided by externally imposed factors (Markland & Tobin, 2004). On the basis of this construct, we expect the effect of fatburning exercise on immediate post-exercise food intake to be stronger in individuals who self-impose physical activity (reflected in lower behavioral regulation scores, i.e., higher amotivation and external regulation; lower introjection, identification, and intrinsic regulation).

The degree to which individuals self-impose physical activity should also be reflected in their subjective experience of exercising. Individuals who find physical activity stressful, tiring, and unenjoyable are predicted to be more likely to overconsume in the fat-burning label condition, because for these individuals exercising means exerting self-control and results in the perception that a goal has been achieved, which makes them more vulnerable to opposing goals, such as eating food after exercise (Fishbach & Dhar, 2005; Fishbach & Shah, 2006). If the exercise bout is labeled fatburning, the label provides a signal that licenses higher postexercise consumption. We therefore predict that the increase in food intake will be highest in individuals who find physical activity stressful and tiring rather than an enjoyable activity that increases wellbeing. Methods Design and participants A laboratory experiment was conducted to test our hypotheses. The study employed a one-factorial design with the label given to an exercise bout (fat-burning exercise vs. endurance exercise) as a between-subjects factor. Participants were randomly assigned to either the fat-burning label condition (with reference to burning off fat) or the endurance label condition (with reference to endurance; the term endurance is central to the physical activity guidelines developed by the American College of Sports Medicine (Haskell et al., 2007) and was therefore chosen to be contrasted with fat-burning). Ninety-six individuals (45 women; overall mean age 26.1 years, SD = 9.4) participated in the study. Participants were recruited on the campus of a large university and consisted of undergraduate and graduate students and university employees. Procedure The study was conducted in the laboratories of a university from Monday through Friday, between 2 pm and 6 pm. Individual appointments were made with the participants to ensure that food consumption was not influenced by the presence of other participants. Participants were not informed about the goal of the study when they were recruited; they were instead told that they would participate in a market research study to evaluate newly developed training software for bicycle ergometers. After the participants had given written informed consent for participation they were equipped with heart rate monitors and seated on the bicycle ergometer. Participants then completed a twenty-minute low-to-moderate intensity bicycle ergometer ride. The heart rate monitor ensured that participants exercised at the same intensity in both experimental conditions. The intensity of the workout was equivalent to a workload of 55–65% of the estimated VO2 max of the participant. During the exercise bout, the label given to the bout was made salient for participants via a poster displayed on the wall in front of participants and on the screen of the bicycle ergometers. In the fatburning label condition, the poster showed the following sentence: ‘fat-burning exercise – developing training software for exercise in the fat-burning zone.’ In the endurance label condition, the control condition, the poster showed the following statement: ‘endurance exercise – developing training software for exercise in the endurance zone.’ When the participants had finished the twenty-minute bicycle ergometer ride, they were told that they could help themselves to drinks and food whilst completing a survey. Behind a partition wall, water bottles and a two-liter bowl filled with Snyder’s of Hanover Pretzel Pieces were arranged so that participants were out of sight

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131

ARTICLE IN PRESS N. Fenzl et al./Appetite ■■ (2014) ■■–■■

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

of the experimenter. Participants could use a spoon to serve themselves pretzels in a smaller bowl (250 g capacity) before seating themselves at a table that was out of sight of the experimenter. Participants then completed a written questionnaire that included some distraction items and sociodemographics (first part of the survey), as well as the moderators and confounding variables (second part of the survey; see Measures). When the participants had finished the questionnaire, they were asked to see the experimenter again. The participants were then thanked for their participation and fully debriefed before they were released. The amount of food eaten was measured using a balance (pre vs. post), subtracting any leftovers. The amount of food eaten (measured in grams) was used as the dependent variable in the following analyses. Measures As a manipulation check participants were asked what percentage of energy they thought they had burned as fat and carbohydrates during the exercise bout. This allowed us to assess whether participants in the fat-burning label condition thought that they had burned a higher proportion of fat than participants in the endurance label condition although they had exercised at the same intensity and for the same time. Behavioral regulation was measured using the revised Behavioral Regulation in Exercise Questionnaire (BREQ-2; Markland & Tobin, 2004). It consists of items relating to intrinsic, identified, and introjected regulation of exercise (e.g., “I exercise because it’s fun,” “I exercise because I value the benefits of exercise,” and “I exercise because I feel guilty when I don’t exercise”); items relating to external regulation of exercise (e.g., “I exercise because other people say I should,” reverse coded); and items relating to amotivation (e.g., “I can’t see why I should bother exercising,” reverse coded). The 18 items were rated on a five-point scale, ranging from 1 = “not at all true for me” to 5 = “very true for me.” The items were averaged to give a single score; the lower the score, the more participants selfimposed physical activity. The reliability of the scale was satisfactory (Cronbach’s α = .71). The Subjective Exercise Experience Scale (SEES) was used to assess how participants experienced the exercise bout (McAuley & Courneya, 1994). It comprises subscales measuring three dimensions (via four items each): positive well-being, psychological distress, and fatigue. Each dimension was measured by asking the participant rate the extent to which four adjectives described their current state on a seven-point scale, ranging from 1 = “not at all” to 7 = “very much.” Items include ‘great’ for positive well-being, ‘miserable’ for psychological distress, and ‘exhausted’ for fatigue. An average score was calculated for each subscale. The scales showed satisfying reliabilities (Cronbach’s α = .81 for positive well-being, Cronbach’s α = .90 for psychological distress, and Cronbach’s α = .87 for fatigue). The following confounding variables were assessed in the survey: self-reported weight and height (to calculate body mass index [BMI]), perceived hunger, perceived tastiness of the food, perceived healthiness of the food, dietary restrained eating behavior, and nutritional knowledge. Perceived hunger was measured via two single items: “How hungry have you been before taking part in this study?” (1 = “not at all hungry,” 5 = “very hungry”) and “When did you have your last meal before you participated in this study?” (assessed in hours). Perceived tastiness was measured via a single item (“This food tastes very good,” anchored at 1 = “strongly disagree” and 5 = “strongly agree”). Perceived healthiness was assessed via a single item (“To me, this product looks very nutritious,” anchored at 1 = “strongly disagree” and 5 = “strongly agree”). Dietary restrained eating behavior was measured using the 10-item Restraint Scale (α = .74) (Herman & Polivy, 1975). Lastly, nutritional knowledge was

3

assessed using Moorman, Diehl, Brinberg, and Kidwell’s (2004) scale that consists of fifteen questions about dietary guidelines and health consequences (0 = lowest score, 15 = highest score).

67 68 69 70 71 Data analyses 72 73 All statistical analyses were conducted by using the software 74 program SPSS/PC+ (version 21, SPSS, Inc., Chicago, IL, USA). Mod75 erated regression analyses were performed to assess the influence 76 of the experimental manipulation on food intake depending on be77 havioral regulation in exercise and subjective exercise experience 78 scores, respectively. Food consumed (in grams) was the depen79 dent variable; the experimental manipulation (0 = fat-burning label, 1 = endurance label), mean-centered moderators (i.e., behavioral reg80 ulation in exercise, positive well-being, psychological distress, and 81 fatigue), and their interaction were modeled as independent vari82 ables. The level of significance was p < .05. 83 84 Results 85 86 Participants spent on average 96 calories (SD = 18) during the 87 twenty-minute bicycle ride, as estimated by heart rate monitor88 ing, and they consumed on average 28 g of pretzels (SD = 30), cor89 responding to an energy intake of 137 calories.1 We can therefore Q2 90 state that the participants (referring to the sample average) tended 91 to overcompensate when given the opportunity to eat food imme92 diately after a low-to-moderate intensity training session. The over93 consumption was equivalent to 41 calories (i.e., 43% more calories 94 compared with calorie expenditure [range: 37–96%]). Moderated re95 gression analyses assessed whether the label given to the exercise 96 bout (fat-burning or endurance) influenced the amount of food eaten 97 after the exercise bout depending on behavioral regulation in ex98 ercise and subjective exercise experience. 99 Prior to the analyses, we conducted a manipulation check. The 100 manipulation check showed that participants who exercised in the 101 fat-burning label condition stated that they had burned a higher per102 centage of fat than participants who exercised in the endurance label 103 condition even though neither exercise intensity nor duration dif104 fered between the groups (M Fat-burning = 41.5%, SD = 24.6% vs. 105 MEndurance = 23.4%, SD = 20.9%, t(89) = 3.78, p < .001).1 The opposite was 106 true for perceived percentage of carbohydrates burned 107 (MFat-burning = 58.5%, SD = 24.6% vs. MEndurance = 76.6%, SD = 20.9%, t(89) = 108 –3.78, p < .001).2 We therefore concluded that the experimental ma109 nipulation had been successful. 110 Next, we assessed whether there were differences between the 111 experimental groups as regards confounding variables. Chi square 112 tests and t-tests showed that there were no significant differences 113 between the experimental groups in terms of gender (χ2(1) = 1.48, 114 p = .22), age (MFat-burning = 24.8, SD = 9.6 vs. MEndurance = 27.4, SD = 9.0, 115 t(94) = –1.35, p = .18), BMI (MFat-burning = 22.5, SD = 2.9 vs. MEndurance = 22.5, 116 SD = 2.6, t(94) = 0.04, p = .97), perceived hunger (rating: 117 MFat-burning = 2.4, SD = 1.5 vs. MEndurance = 2.7, SD = 2.0, t(94) = –0.83, p = .41; 118 hours since last meal: MFat-burning = 3.3, SD = 1.9 vs. MEndurance = 3.6, 119 SD = 2.0, t(94) = –0.75, p = .46), perceived tastiness of the food 120 (MFat-burning = 3.7, SD = 1.0 vs. MEndurance = 3.5, SD = 1.1, t(94) = 0.67, p = .50), 121 perceived healthiness of the food (MFat-burning = 2.4, SD = 1.0 vs. 122 MEndurance = 2.4, SD = 1.2, t(94) = 0.07, p = .83), dietary restrained eating 123 124 125 1 Participants in the fat-burning condition on average consumed 27 g (SD = 27) and 126 participants in the endurance condition on average consumed 30 g (SD = 32) of the 127 pretzels. The following moderated regression analyses revealed that there was no 128 significant main effect of the labeling of the exercise bout. 129 2 There were missing values for five participants. We did not exclude these par130 ticipants, because they successfully completed the exercise trial and filled in the ques131 tionnaire completely (except for estimations on macronutrient composition of their 132 energy expenditure). 133

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

ARTICLE IN PRESS 4

1 2 3 4

N. Fenzl et al./Appetite ■■ (2014) ■■–■■

Fig. 1. Influence of labeling an exercise bout fat-burning (vs. endurance) on postexercise food consumption depending on individuals’ behavioral regulation in exercise. Notes. The Johnson-Neyman points depict the region of significance (i.e., below values of 3.86 and above values of 4.32).

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

behavior (MFat-burning = 11.2, SD = 6.1 vs. MEndurance = 11.7, SD = 5.8, t(94) = –0.36, p = .72), and nutritional knowledge (MFat-burning = 7.4, SD = 2.9 vs. MEndurance = 7.9, SD = 2.9, t(94) = –0.85, p = .39). None of these confounding variables affected any of the hypothesized relationships, as indicated by non-significant beta coefficients of the interaction effects with the experimental manipulation in moderated regression analyses (p’s > .05). Therefore, and for reasons of clarity, we present the results of the moderated regression model analyses without confounding variables. To find out whether food consumption in the two conditions was influenced by the degree to which individuals self-impose physical activity, moderated regression analyses were conducted. In the first moderated regression analysis, food consumed (in grams) was the dependent variable3 and the experimental manipulation (0 = fatburning label, 1 = endurance label), behavioral regulation in exercise (mean-centered), and their interaction were modeled as independent variables. The results showed that higher behavioral regulation in exercise (indicating a lower tendency to self-impose physical activity) was negatively associated with food intake (b = – 39.3, SE = 9.8, p < .001). There was no significant effect of the exercise labeling on food intake (b = 4.2, SE = 5.5, p = .45). More importantly, the interaction between labeling and behavioral regulation in exercise was significant (b = 77.6, SE = 14.8, p < .001; see Q3 Fig. 1). The interaction can be described using the Johnson-Neyman technique (Hayes & Matthes, 2009). As seen in Fig. 1, labeling the exercise bout fat-burning exercise increased food intake for behavioral regulation scores below values 3.86. The lowest behavioral regulation in exercise score in our sample was 2.81. This result is in accordance with our predictions: participants with low behavioral regulation scores consumed more food after the exercise bout if it was labeled fat-burning exercise than in the endurance exercise condition. The reverse was true for behavioral regulation scores above 4.32, indicating that participants with high behavioral regulation scores consumed less food after they had cycled in the fat-burning label condition than in the endurance exercise condition. We had not predicted this relationship; however, it indicates that high selfdetermination may protect consumers against overcompensatory

3 Here, and in the following analyses, we used original scores (i.e., food consumption in grams). Moderated regression analyses using the square root of this variable (accounting for skewness, s = 2.38 for original scores [s = 0.37 for square root scores]) as dependent variable produced similar results. For ease of interpretation, we used original scores.

food consumption after the exercise bout in the fat-burning label condition. We assessed the subjective experience of exercising as another facet of the degree to which individuals self-impose physical activity. This allowed us to find out whether the relationship between behavioral regulation in exercise and post-exercise food consumption could be replicated using the SEES; self-imposed exercisers were expected to be less likely to experience the activity as enjoyable and more likely to experience the activity as stressful and tiring. This prediction was confirmed, because all three SEES dimensions correlated significantly with BREQ-2 scores in the expected directions (r = .25, p < .05 for positive well-being, r = –.38, p < .001 for psychological distress, and r = –.30, p < .01 for fatigue). Accordingly, we further hypothesized that individuals who perceived the exercise bout as unenjoyable, stressful, and tiring would eat more food after the exercise bout in the fat-burning label condition than individuals who perceived the exercise as enjoyable, but not stressful or tiring. To assess these relationships, we conducted three moderated regression analyses using variables as above except for the moderators and their interaction. The latent variables were mean-centered prior to the analyses. The results showed that there was a nonsignificant effect of positive well-being on food intake (b = –5.2, SE = 4.9, p = .29) and a non-significant effect of labeling on food intake (b = 4.3, SE = 6.0, p = .48). More importantly, the interaction between the exercise label and positive well-being was significant and positive (b = 17.0, SE = 6.2, p < .01). Using the Johnson-Neyman technique to describe the interaction effect, the results showed that the effect was significant for positive well-being scores between 2.00 and 3.36. As seen in Fig. 2 (top), labeling the exercise bout fatburning increased food intake for this range of scores. The lowest positive well-being score in our sample was 2.00. The reverse effect was observed for positive well-being scores above 5.58. Although we had not predicted this latter result, it is in line with the finding that labeling an exercise bout fat-burning rather than endurance produced lower post-exercise food intake in participants with high behavioral regulation in exercise. Psychological distress was positively associated with food intake (b = 14.1, SE = 6.9, p < .05) and exercise labeling was not associated with food intake (b = 2.6, SE = 6.0, p = .67). More importantly, the interaction between the exercise label and psychological distress was significant (b = –20.1, SE = 7.9, p < .05). As shown in Fig. 2 (middle), the significance level at low levels of psychological distress ranged from distress values of 1.00 to values of 1.12. Again, this finding had not been predicted, but is line with results for behavioral regulation in exercise and positive well-being. As hypothesized, food intake was significantly higher when the exercise bout was labeled fatburning rather than endurance at high levels of psychological distress, that is, scores between 3.58 and 6.80 (the latter being the highest score in our sample). The results therefore support our prediction: participants who found the exercise bout more stressful consumed more food afterwards when it was labeled fat-burning exercise rather than endurance exercise. We also conducted a moderated regression analysis using fatigue as independent variable and the interaction between the exercise label and fatigue as the moderator. The results showed that there was a positive association between fatigue and food intake (b = 26.5, SE = 8.2, p < .01) and a non-significant effect of labeling on food intake (b = 1.8, SE = 5.8, p = .75). As hypothesized, the interaction effect between the exercise label and fatigue was significant (b = –32.0, SE = 9.2, p < .001). The calculation of ranges of significance showed that food intake was higher in the fat-burning label condition for fatigue score above 1.84. The highest score reported in our sample was 7.00. The results thus supported our hypothesis. There was no reverse effect in participants with low fatigue scores. Fig. 2 (bottom) illustrates the findings.

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115

ARTICLE IN PRESS N. Fenzl et al./Appetite ■■ (2014) ■■–■■

1 2 3 4 5 6 7

Fig. 2. Influence of labeling an exercise bout fat-burning (vs. endurance) on postexercise food consumption depending on individuals’ subjective exercise experience, including positive well-being (top), psychological distress (middle), and fatigue (bottom). Notes. The Johnson-Neyman points depict the region of significance (i.e., below values of 3.36 and above values of 5.58 for positive well-being; below values of 1.12 and above values of 3.58 for psychological distress; above values of 1.84 for fatigue).

8 9 10 11 12 13 14 15 16 17

Discussion The study showed that self-imposed exercisers – individuals with low behavioral regulation and high psychological distress, high fatigue levels, and low positive well-being when exercising – ate more food immediately after an exercise bout when the bout was labeled fat-burning exercise rather than endurance exercise. Although food consumption is not necessarily associated with negative health consequences, there is increasing evidence that

5

individuals in developed countries struggle to maintain their energy balance, tending to consume more energy than they expend. In the long run, a positive energy balance causes weight gain, and weight gain is associated with several health risks (Pedersen, 2013). In this study, participants overcompensated for the energy spent via immediate post-exercise food intake (by 43% when transferred to energy equivalents). The label given to the exercise bout (fat-burning or endurance) did not increase the amount of food eaten after the exercise bout per se, because the main effect of labeling was not significant. However, the direction of the effect depended on the degree to which individuals self-imposed physical activity, indexed by measures of behavioral regulation in exercise (Markland & Tobin, 2004) and subjective exercise experience (McAuley & Courneya, 1994). Behavioral regulation in exercise is the degree of self-determination of individuals with respect to exercise. At higher levels of behavioral regulation, self-determination is higher. High behavioral regulation characterizes an individual who has internalized the regulation of their behavior; motivation to exercise emanates from the self rather than from external pressures. Low behavioral regulation characterizes individuals who rely on external motivating factors to regulate their behavior (Markland & Tobin, 2004). We found that low regulators ate more food after the exercise bout when it was labeled fat-burning exercise. The fat-burning prime may have activated associations related to fat-burning exercise, including social knowledge structures, and this probably influenced post-exercise eating behavior (Bargh, 2006). These individuals then considered physical activity legitimized rewarding activities such as eating tasty food (Berthoud, 2004). The study also provided some preliminary evidence that labeling an exercise bout fat-burning (vs. endurance) may reduce consumption in individuals with high behavioral regulation. We had not predicted this finding. Two potential mechanisms seem possible: participants may have been reminded that they have not achieved their long-term goals yet (i.e. to burn off body fat), and that they should remain committed to that goal; they may have eaten less food in order to stimulate their fat metabolism further (Fishbach & Dhar, 2005). Another potential explanation is that they may have felt that their self-determination had been challenged and that they needed to exercise longer in order to reward themselves, because exercise is an instrinsically enjoyable and rewarding activity. In contrast to participants with low behavioral regulation, they may not have perceived the food as a reward. Also, individuals with high behavioral regulation scores who exercised in the endurance exercise condition may have thought that they worked hard to achieve their goal of staying fit (Fishbach et al., 2006) and therefore felt licensed to reward themselves; this potentially explains higher consumption volumes in these participants in the endurance (vs. fatburning) exercise condition. Future research is needed to show which of the mechanisms explains these effects best. The finding that emotional states during exercising can influence post-exercise food consumption in certain contexts is important, because there are individual differences in the affective response to exercise (Van Landuyt, Ekkekakis, Hall, & Petruzzello, 2000). Individuals who perceived physical activity as stressful and unenjoyable were more likely to overcompensate in the fat-burning label condition, whereas the reverse was true for individuals who perceived physical activity as enjoyable and unstressful. For fatigue scores, our results support the hypothesis that labeling an exercise bout fat-burning increases post-exercise food consumption in individuals who state that exercising makes them tired. We did not find the reverse relationship in individuals with low fatigue levels. The behavior of individuals who found physical activity stressful and tiring rather than enjoyable was in line with our assumptions: for these participants, the salience of the burning-fat concept may have made them perceive that the goal of burning calories has been

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

ARTICLE IN PRESS N. Fenzl et al./Appetite ■■ (2014) ■■–■■

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

achieved, which makes them more vulnerable to opposing goals, such as eating food after exercise (Fishbach & Dhar, 2005; Fishbach & Shah, 2006). Individuals who found physical activity enjoyable and unstressful ate more food in the endurance label condition than in the fatburning label condition. For them, the salience of the enduranceexercise concept may have triggered the feeling that they worked hard and used their muscle resources; this may have reduced their monitoring of self-control and increased the likelihood of indulgent choices in a subsequent task (Baumeister, Bratslavsky, Muraven, & Tice, 1998). Future research is needed to find out whether the concept of self-control helps explain the behavior of individuals who perceive physical activity as enjoyable and unstressful. Implications Fat-burning programs and fat-burning classes are targeted at individuals who are supposed to benefit most from exercising at a lowto-moderate intensity. This often includes individuals who have just started an exercise regime (because their bodies are not used to exercising at higher intensities), individuals who have been advised to avoid high-intensity exercise for health reasons such as chronic diseases and obesity, and individuals who want to lose weight and think that the best way to reach this goal is by exercising at an intensity where the percentage of fat burned is highest (Biddle, Petrolini, & Pearson, 2013). This last group of individuals is particularly interested in inducing a negative energy balance via physical activity. Low-to-moderate intensity exercise is typically recommended for individuals who want to prevent or reduce overweight and obesity (Thompson, Townsend, Boughey, Patterson, & Bassett, 1998), and our results indicate that labeling an exercise bout fat-burning may have a negative effect, in the form of an overcompensatory increase in food intake immediately after the exercise bout, in individuals who rely on external motivation to carry out physical activity. Self-imposing physical activity is a phenomenon that is often observed in individuals with body weight problems (Donovan & Penny, 2014). Giving exercise programs and classes labels other than fat-burning (i.e., names that do not have a close association with perceived goal fulfillment) or making labels such as fatburning less salient during exercise may help reduce the overcompensation effect on immediate post-exercise food intake in these individuals. In addition to changing the labels given to exercise programs and classes and making them less salient, one might also recommend that health professionals in the field of exercise and nutrition attempt to strengthen the self-determination of individuals who are at greatest risk of being misled by the use of the fat-burning label for certain exercise bouts. In the exercise domain, individuals are often told what is safe to do for them (and what is not safe), how hard and for how long they should work out; this is particularly true in exercise for weight management (Haskell et al., 2007). The results of our study suggest that individuals should learn to consider physical activity, when framed as fat-burning exercise, as a rewarding and enjoyable activity and increase their ability to use internal factors to regulate behavior rather than relying on external, including social and societal motivating factors and pressures. Choosing a sport that matches the interest of the individual and which he or she will find enjoyable is one strategy that may help to achieve this goal. Manipulating how physical activity is framed can also increase self-determination. Werle et al. (2011 p. 335) stated that, “instead of describing [exercise] as a facilitation to weight loss, it can be presented as a critical way to tone one’s self, strengthen bones, improve posture, and so forth.” Use of such messages would emphasize that individuals make an active contribution to their health and wellbeing when they exercise, and that they determine their behavior and the resulting consequences by themselves. Van Kleef, Shimizu,

and Wansink (2011) reported overcompensation and reduced motivation to exercise if individuals feel under pressure to do so or feel an obligation to exercise and to be fit and healthy. Service providers could improve the framing of physical activity – and perhaps avoid these effects – by including messages that increase selfdetermination and positive well-being and reduce stress and fatigue (Hills & Byrne, 2004). The results of our study, however, indicate that individuals who do not self-impose physical activity also respond to the labeling of exercise bouts as regards post-exercise food consumption: individuals with high behavioral regulation and individuals who perceived physical activity as enjoyable and unstressful ate more food after the exercise bout when the bout was labeled endurance (vs. fat-burning) exercise, potentially because the labeling triggered associations with the concept of endurance. Practitioners may take this into account when designing target-group specific lifestyle interventions. Another recommendation for both high and low regulators is that individuals should learn to use physiological signals to infer the intensity of their exercise program rather than relying on external cues such as the label given to an exercise program. We have argued that the labels activate automatic associations; the effect of such implicit associations may go beyond direct effects such as promoting understanding of exercise intensities and the ability to interpret exercise intensities and estimate calorie expenditure in exercise. Previous research has shown that promoting explicit knowledge is a difficult task in the exercise domain (Braham, Rosenberg, & Begley, 2012). This is particularly true when health practitioners prescribe physical activity (which is typically an indicator that the individual relies on external factors to carry out exercise); in these circumstances warning patients or clients about compensation is advisable, but may not be as effective as one might wish (King, 1999). Hills and Byrne (2004, p. 316) referred to previous research in their field (Hills & Byrne, 1998) when they wrote that “exercise prescription is both an art and a science.” In this study participants in the fat-burning label condition believed that a higher percentage of energy expenditure was fat burning than actually is the case for regular low-to-moderate exercise. We therefore recommend that practitioners make physical activity prescriptions in such a way that individuals are not tempted to feel – albeit implicitly – that they have attained desirable long-term goals such as attractive body appearance and low body weight because of the label attached to the form of exercise they have undertaken.

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

Study limitations This study has several limitations. First, the generalizability of findings obtained in laboratory settings, particular those related to measurement of food intake, has been questioned (Meiselman, 1992). We cannot rule out the possibility that participants’ post-exercise food intake in our study may have been higher or lower than in reallife situations, for example, when leaving the fitness club. Our sample of university students and employees was rather homogeneous and so the results are not representative of the general population. Second, macronutrient composition and perception of the food may have influenced the eating behavior of our participants (Knight & Boland, 1989). The food used in this study had a fat content of 23.6 g per 100 g (equivalent to 33.7% of the calories consumed). Further research may find out whether post-exercise food intake in individuals who have exercised in the fat-burning zone differs according to fat content of the food. The tendency for overcompensation might be even higher when high-fat food is made available to individuals who rely on external factors to carry out exercise, because high-fat food is perceived as more rewarding and enjoyable than low-fat food (Raghunathan, Naylor, & Hoyer, 2006).

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131

ARTICLE IN PRESS N. Fenzl et al./Appetite ■■ (2014) ■■–■■

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

The potential effects of the social desirability bias are another limitation. In particular, social desirability may have affected the participants’ responses on the self-report instruments that we administered, resulting in for example a tendency to evaluate physical activity more positively in terms of SEES, that is, giving higher positive well-being scores and lower psychological distress and fatigue scores, because this is desirable in our society. Future research could manipulate rather than measure these variables, to gain further insight into the relevance of these indicators of the tendency to rely on external motivating factors to carry out exercise to post-exercise energy compensation. Concluding remarks Previous studies have looked at how low-fat claims on foods influence consumption behavior. This study is the first to show that fat-burning claims about physical activity interact with food consumption. Although this study’s goal was not to challenge existing guidelines and recommendations relating to the intensity at which individuals should exercise, it highlights the relevance of the label given to exercise programs when exercising at low-to-moderate intensity. Labeling an exercise bout fat-burning increased food consumption after the exercise bout in individuals who self-imposed physical activity. The label likely provided an implicit cue to these participants that made them overcompensate for the energy expended. We hope that this result inspires future research into the effects of contextual and psychological factors on energy balance effects. References Bargh, J. A. (2006). What have we been priming all these years? On the development, mechanisms, and ecology of nonconscious social behavior. European Journal of Social Psychology, 36(2), 147–168. Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion. Is the active self a limited resource? Journal of Personality and Social Psychology, 74(5), 1252–1265. Berthoud, H.-R. (2004). Mind versus metabolism in the control of food intake and energy balance. Physiology and Behavior, 81(5), 781–793. Biddle, S. J., Petrolini, I., & Pearson, N. (2013). Interventions designed to reduce sedentary behaviours in young people. A review of reviews. British Journal of Sports Medicine, 48(3), 182–186. Blundell, J. E., & King, N. A. (1999). Physical activity and regulation of food intake. Current evidence. Medicine and Science in Sports and Exercise, 31(11), 573–583. Braham, R., Rosenberg, M., & Begley, B. (2012). Can we teach moderate intensity activity? Adult perception of moderate intensity walking. Journal of Science and Medicine in Sport, 15(4), 322–326. Donovan, C. L., & Penny, R. (2014). In control of weight. The relationship between facets of control and weight restriction. Eating Behaviors, 15(1), 144–150. Eriksson, J., Taimela, S., & Koivisto, V. A. (1997). Exercise and the metabolic syndrome. Diabetologia, 40(2), 125–135. Fishbach, A., & Dhar, R. (2005). Goals as excuses or guides. The liberating effect of perceived goal progress on choice. The Journal of Consumer Research, 32(3), 370–377. Fishbach, A., Dhar, R., & Zhang, Y. (2006). Subgoals as substitutes or complements. The role of goal accessibility. Journal of Personality and Social Psychology, 91(2), 232–242. Fishbach, A., & Shah, J. Y. (2006). Self-control in action. Implicit dispositions toward goals and away from temptations. Journal of Personality and Social Psychology, 90(5), 820–832. Fitzsimons, G. M., Chartrand, T. L., & Fitzsimons, G. J. (2008). Automatic effects of brand exposure on motivated behavior. How apple makes you “think different”. The Journal of Consumer Research, 35(1), 21–35.

7

Geyskens, K., Pandelaere, M., Dewitte, S., & Warlop, L. (2007). The backdoor to overconsumption. The effect of associating “low-fat” food with health references. Journal of Public Policy and Marketing, 26(1), 118–125. Hall, K. D., Heymsfield, S. B., Kemnitz, J. W., Klein, S., Schoeller, D. A., & Speakman, J. R. (2012). Energy balance and its components. Implications for body weight regulation. The American Journal of Clinical Nutrition, 95(4), 989–994. Haskell, W. L., Lee, I., Pate, R. R., Powell, K. E., Blair, S. N., Franklin, B. A., et al. (2007). Physical activity and public health. Updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Medicine and Science in Sports and Exercise, 39(8), 1423–1434. Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression. SPSS and SAS implementations. Behavior Research Methods, 41(3), 924–936. Herman, C. P., & Polivy, J. (1975). Anxiety, restraint, and eating behavior. Journal of Abnormal Psychology, 84(6), 666–672. Hills, A. P., & Byrne, N. M. (1998). Exercise prescription for weight management. The Proceedings of the Nutrition Society, 57(1), 93–103. Hills, A. P., & Byrne, N. M. (2004). Physical activity in the management of obesity. Clinics in Dermatology, 22(4), 315–318. King, N. A. (1999). What processes are involved in the appetite response to moderate increases in exercise-induced energy expenditure? The Proceedings of the Nutrition Society, 58(1), 107–113. King, N. A., Tremblay, A., & Blundell, J. E. (1997). Effects of exercise on appetite control. Implications for energy balance. Medicine and Science in Sports and Exercise, 29(8), 1076–1089. Knight, L. J., & Boland, F. J. (1989). Restrained eating. An experimental disentanglement of the disinhibiting variables of perceived calories and food type. Journal of Abnormal Psychology, 98(4), 412–420. Koenigstorfer, J., Groeppel-Klein, A., Kettenbaum, M., & Klicker, K. (2013). Eat fit. Get big? How fitness cues influence food consumption volumes. Appetite, 65, 165–169. Markland, D., & Tobin, V. (2004). A modification to the behavioural regulation in exercise questionnaire to include an assessment of amotivation. Journal of Sport and Exercise Psychology, 26(2), 191–196. McAuley, E., & Courneya, K. S. (1994). The subjective exercise experiences scale (SEES). Development and preliminary validation. Journal of Sport and Exercise Psychology, 16, 163–177. Meiselman, H. L. (1992). Methodology and theory in human eating research. Appetite, 19(1), 49–55. Moorman, C., Diehl, K., Brinberg, D., & Kidwell, B. (2004). Subjective knowledge, search locations, and consumer choice. The Journal of Consumer Research, 31(3), 673–680. Pedersen, S. D. (2013). Metabolic complications of obesity. Clinical Endocrinology and Metabolism, 27(2), 179–193. Raghunathan, R., Naylor, R. W., & Hoyer, W. D. (2006). The Unhealthy = Tasty intuition and its effects on taste inferences, enjoyment, and choice of food products. Journal of Marketing, 70(4), 170–184. Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. The American Psychologist, 55(1), 68–78. Schubert, M. M., Desbrow, B., Sabapathy, S., & Leveritt, M. (2013). Acute exercise and subsequent energy intake. A meta-analysis. Appetite, 63, 92–104. Serdula, M. K., Mokdad, A. H., Williamson, D. F., Galuska, D. A., Mendlein, J. M., & Heath, G. W. (1999). Prevalence of attempting weight loss and strategies for controlling weight. Journal of the American Medical Association, 282(14), 1353–1358. Thompson, D. L., Townsend, K. M., Boughey, R., Patterson, K., & Bassett, D. R., Jr. (1998). Substrate use during and following moderate- and low-intensity exercise. Implications for weight control. European Journal of Applied Physiology and Occupational Physiology, 78(1), 43–49. Ueda, S., Yoshikawa, T., Katsura, Y., Usui, T., Nakao, H., & Fujimoto, S. (2009). Changes in gut hormone levels and negative energy balance during aerobic exercise in obese young males. The Journal of Endocrinology, 201(1), 151–159. Van Kleef, E., Shimizu, M., & Wansink, B. (2011). Food compensation. Do exercise ads change food intake? International Journal of Behavioral Nutrition and Physical Activity, 8, 661–664. Van Landuyt, L. M., Ekkekakis, P., Hall, E. E., & Petruzzello, S. J. (2000). Throwing the mountains into the lakes. On the perils of nomothetic conceptions of the exercise-affect relationship. Journal of Sport and Exercise Psychology, 22, 208–234. Wansink, B., & Chandon, P. (2006). Can “low-fat” nutrition labels lead to obesity? Journal of Marketing Research, 43(4), 605–617. Warburton, D. E., Nicol, C. W., & Bredin, S. S. (2006). Health benefits of physical activity. The evidence. Canadian Medical Association Journal, 174(6), 801–809. Werle, C. O., Wansink, B., & Payne, C. R. (2011). Just thinking about exercise makes me serve more food. Physical activity and calorie compensation. Appetite, 56(2), 332–335.

Please cite this article in press as: Navina Fenzl, Katja Bartsch, Joerg Koenigstorfer, Labeling exercise fat-burning increases post-exercise food consumption in selfimposed exercisers, Appetite (2014), doi: 10.1016/j.appet.2014.05.030

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136