Acute aerobic exercise increases implicit approach motivation for dessert images

Abstract

We examined the effect of acute exercise compared to a cognitive task on implicit approach/avoidance motivation to dessert food images using the Dessert–Approach–Avoidance Task. Participants randomized to exercise had a greater increase in approach motivation to dessert images compared to those completing cognitive tasks (p=0.046), adjusting for disordered eating, task difficulty, and changes in negative affect. This study provides the first evidence for the use of the Dessert–Approach–Avoidance Task to evaluate the effects of acute exercise on implicit motivations for dessert images. Future studies should examine implicit response to food images using the Dessert–Approach–Avoidance Task in response to chronic exercise.

Keywords

activity diet eating behavior exercise exercise behavior obesity

Introduction

College students have the opportunity to develop healthy habits, including food choices, which they can carry throughout their entire lives (Colic Baric et al., 2003). Both decreased physical activity and increased consumption of unhealthy energy-dense foods are generally accepted as major contributors to the steady rise in obesity (World Health Organization, 2003).

Overall, only 49.9 percent of non-varsity athlete college students meet the current recommendations for moderate-to-vigorous intensity physical activity (MVPA) (Dinger et al., 2014). Given that meeting the current MVPA recommendation is also associated with adequate daily fruit and vegetable consumption and other health behaviors (Dinger et al., 2014), it is important to promote physical activity in this population.

Exercise may affect the palatability of food and may alter the hedonic reward of particular nutrients (Elder and Roberts, 2007). For example, research has shown that adults who took a 5-minute brisk walk had a reduced urge to snack, providing evidence that moderate exercise may help with self-regulatory behaviors (Thayer et al., 1993). Although many individuals decrease their desire for unhealthy foods after exercising, some individuals see unhealthy foods as desirable after a workout (Finlayson et al., 2011). Many factors play a role in whether individuals view exercise as a means to justify eating high-calorie foods or exercise as a way to reduce consumption of high-calorie foods, including personality differences, personal exercise motivations, or weight loss/maintenance goals (Lowry et al., 2000; Ryan et al., 1997). These are more explicit attitudes and motivations toward food that are influenced by our cognitions. Implicit food motivations reflect bottom-up processes that underlie our behavior and are not influenced by our personality differences, goals, or attitudes. No previous studies have explored the relationship between exercise and implicit food motivations.

Studies in behavioral therapy and alcoholism have utilized the Approach–Avoidance Task (AAT) (Rinck and Becker, 2007; Wiers et al., 2009, 2010), a joystick task that measures action tendencies in response to emotional stimuli. The joystick task assesses one’s automatic approach/withdraw biases toward assorted stimuli (Cacioppo et al., 1993). In this task, arm flexion (pulling the joystick toward oneself) is associated with more positive evaluations than extension (pushing the joystick away) (Cacioppo et al., 1993). More recent research has replicated findings that individuals exhibit faster withdrawal (extension/push) movements in response to fearful stimuli (e.g. spiders for arachnophobics) (Rinck and Becker, 2007) and faster approach (flexion/pull) movements in response to appetitive stimuli (e.g. alcohol for hazardous drinkers) (Wiers et al., 2010). The AAT has also been used in intervention research with hazardous drinkers and alcoholics (Wiers et al., 2009, 2011). Participants were implicitly “trained” to avoid alcoholic beverages, such that the majority of the time participants pushed the joystick away in response to the way images of alcoholic beverages were tilted (Wiers et al., 2009, 2011). Participants implicitly trained to avoid the alcoholic beverages drank less alcohol immediately after training (Wiers et al., 2009), and alcoholics who had four training sessions had better treatment outcomes at 1-year follow-up (Wiers et al., 2011). These studies using the alcohol-AAT show promising evidence that implicit bias can be altered through a training version of the AAT. Recent research has demonstrated that showing dessert images (Gable and Harmon-Jones, 2008; Juergensen and Demaree, 2015) causes approach-motivated action tendencies (Juergensen and Demaree, 2015), but no study has investigated whether implicit approach motivation for dessert images can be altered. If implicit approach motivation toward dessert images can be altered through exercise, a similar training paradigm could help individuals reduce consumption of dessert foods.

Exercise may result in modifications to approach-motivated action tendencies. Short bouts of exercise, such as 15 minutes of brisk walking, can reduce sensitivity, urges, and consumption for high-caloric foods, including chocolate (Oh and Taylor, 2012; Taylor and Oliver, 2009). Additionally, 60 minutes of cycling, compared to no exercise, reduced neuronal responses to high- and low-energy foods in brain areas associated with pleasure of food, incentive motivation to eat, and anticipation and consumption of food (Evero et al., 2012). Thus, exercise may modulate implicit desire for certain foods, which has important implications for weight loss programs to emphasize the importance of physical activity in reducing unhealthy eating. Although we could measure consumption of a highly palatable food post-exercise, these results would be subject to cognitive biases underlying the decision to consume that food or not. Therefore, we evaluated the potential effects of acute aerobic exercise on implicit approach motivation using the Dessert–Approach–Avoidance Task (D-AAT).

We hypothesized that participants in the exercise condition would experience decreased approach bias toward dessert images relative to neutral images on the D-AAT compared to those in the cognitive task condition. Second, we expected participants with higher levels of self-reported physical activity would have decreased implicit approach bias to the dessert images on the D-AAT compared to individuals with lower levels of physical activity.

Methods

Study participants

Participants included undergraduates from an Introduction to Psychology research pool. A power analysis using G*Power (Faul et al., 2009) revealed that with effect size f = 0.25 (a medium effect), α = 0.05, and power = 0.80, the total sample size needed to include 128 participants. In total, 127 undergraduates volunteered and provided informed written consent. Data from 39 participants were excluded from the study for the following reasons: (1) 26 were non-native English speakers and their desire for American dessert foods is likely different among individuals who are not as acculturated May et al., 2016), (2) 5 likely had an eating disorder according to their score on the Eating Attitudes Test-26 (EAT-26), (3) 3 were missing Time 2 D-AAT data, (4) 3 were missing body fat percentage data, and (5) 2 were missing substantial questionnaire data (1 missing the Time 2 Positive and Negative Affect Schedule (PANAS) and the Level of Perceived Effort/Exertion Scale and 1 missing more than half of the items on the Exercise Motivation Inventory-2 (EMI-2)). The final analytic sample included 88 participants.

Procedures

Participants were informed that they may participate in physical activity and were asked to come to the research lab dressed in comfortable clothing. After consenting, participants’ body composition was measured (see section “Physiological measures”). Participants then completed several questionnaires (see section “Behavioral measures”). We measured both resting heart rate and heart rate during the entire exercise bout or cognitive task. Individuals participated in the study at various start times throughout the day, ranging from 9:00 am to 5:30 pm. Two team members administered the tasks (C.N.M. and D.B.).

Next, participants completed an adapted version of the AAT (Rinck and Becker, 2007; Wiers et al., 2009, 2010), the D-AAT, to provide a measure of the approach or avoidance tendencies elicited by 40 neutral images and 40 dessert images. We assessed PANAS after completion of the D-AAT. Individuals then either participated in the exercise condition or the cognitive task condition for approximately 20 minutes. Participants were randomized to the exercise bout or cognitive task condition by flipping a coin.

Exercise condition

If randomized into the exercise condition, participants rode a stationary upright bicycle (Schwinn 170) for 20 minutes at a moderate intensity, between 55 and 65 percent of their maximum heart rate (Centers for Disease Control and Prevention (CDC), 2011). Maximum heart rate was calculated by 220-age (Fox et al., 1971). After the target heart rate range was calculated, participants began riding the stationary bike. The experimenter left the room and monitored the participant’s effort levels through a one-way mirror. Participants received feedback from a computer screen controlled by the experimenter advising them to increase, decrease, or keep their effort levels consistent to remain in their target heart rate range. There was a brief warm up period (~5 minutes); then, after the participant reached the target heart rate, they completed 20 minutes in the target heart rate zone. This was followed by a brief cool-down session (~5 minutes).

Control/cognitive task condition

Participants in the cognitive task condition completed computerized working memory tasks, the Operation Span and the Reading Span, which lasted approximately 20 minutes. Operation Span (OSPAN) (Turner and Engle, 1989) and Reading Span (RSPAN) (Daneman and Carpenter, 1980) are working memory tasks measuring participants’ ability to maintain information in memory while performing other cognitive tasks.

Primary outcome measure

D-AAT

Using E-PRIME 2.0 Professional (Psychology Software Tools, Pittsburgh, PA), participants completed the D-AAT, an adapted version of the AAT that has been described in more detail previously May et al., 2016; Juergensen and Demaree, 2015). Participants completed 160 test trials where pictures of appetizing desserts and neutral items were displayed equally often in push- or pull-format. Reaction times were measured for responses to all stimuli (in milliseconds). Faster responses to pull movements indicate a more approach-oriented action tendency and faster responses to push movements indicate a more withdraw-oriented action tendency. The AAT assesses automatic/implicit motivational response (approach vs withdraw) much like the Implicit Association Test (IAT) measures automatic/implicit valence response (good vs bad) (Krieglmeyer et al., 2013). The neutral images included everyday items such as a stool, clock, or light bulb chosen from the International Affective Picture System (IAPS) (Center for the Study of Emotion and Attention, 1999) and dessert images included images used in previous studies (Gable and Harmon-Jones, 2008; Juergensen and Demaree, 2015). The neutral items have been rated (1 = low and 9 = high) on the Self-Assessment Manikin (Bradley and Lang, 1994) as neutral on the scales of valence (M = 4.97, standard deviation (SD) = 0.249) and arousal (M = 2.91, SD = 0.585).¹ The dessert images from Gable and Harmon-Jones (2008) included various items, such as chocolate cake, ice cream, and brownies. These dessert images were rated (1 = really desired, 9 = did not desire) significantly more desirable (M = 4.12) than neutral pictures (M = 7.15), indicating that the dessert images evoked approach motivation.

Immediately after completing the exercise or cognitive task, participants rated their perceived level of exertion and completed a second assessment of positive and negative affect (PANAS). Finally, participants completed the D-AAT for a second time.

Physiological measures

Participants’ height and weight information were measured by tape measure and a BalanceFrom digital scale, respectively.

Body fat composition was measured with a bipolar impedance device (Omron Body Fat Analyzer HBF-306C).

Heart rate was measured using a Polar Heart Rate Monitor RCX3 during both the exercise and working memory tasks.

Behavioral measures

Demographic questionnaire

Participants completed a demographic questionnaire that included information about age, gender, race, and ethnicity. Additional questions about dieting, the amount of time since the participant’s last food consumption, and any health conditions that would prevent the participant from participating in physical activity were also included.

Physical Activity Readiness Questionnaire

The Physical Activity Readiness Questionnaire (PAR-Q) is a seven-item self-report measure identifying any risk factor that may prevent individuals from safely participating in physical activity (Canadian Society for Exercise Physiology, 2002).

Satiety Labeled Intensity Magnitude

The Satiety Labeled Intensity Magnitude (SLIM) is a self-report measure used to assess hunger/satiety ranging from “greatest imaginable fullness” to “greatest imaginable hunger” (Cardello et al., 2005). Higher scores reflect greater hunger. The average reliability of the SLIM is 0.90 (Cardello et al., 2005).

PANAS

The PANAS is a 20-item measure used to assess what participants are feeling (both positive and negative) at the time the PANAS is given (Watson et al., 1988). Each item is measured on a 5-point Likert scale from 1 (very slightly or not at all) to 5 (very much). The PANAS has high internal consistency (Cronbach’s α) for both the positive affect items (0.86–0.90) and the negative affect items (0.84–0.87) (Watson et al., 1988).

Level of perceived effort/exertion

Level of perceived effort/exertion was assessed using an in-house questionnaire immediately after participants completed the exercise or working memory tasks. Participants were asked “on a scale from 0 (not difficult) to 9 (extremely difficult) rate the level of difficulty in performing the previous task; on a scale from 0 (no effort) to 9 (extreme effort) rate the level of effort required to perform the previous task; on a scale from 0 (no energy) to 9 (extreme energy) rate the level of energy required for the previous task.” The perceived exertion had good internal consistency (α = 0.87).

Eating Attitudes Test-26

The Eating Attitudes Test-26 (EAT-26) is a 26-item questionnaire used to identify risk for an eating disorder (Garner et al., 1982). Participants responded to each item indicating the frequency of their feelings on a 4-point Likert scale from “Always” to “Never.” The EAT-26 has three subscales: Dieting, Food Preoccupation, and Oral Control. Responses are summed to produce a total score. A total score of 20 or higher has been used to help classify people who are likely experiencing an eating disorder (Garner et al., 1982; King, 1991). Because eating disorders can significantly alter people’s motivation for food (e.g. Schienle et al., 2009), we used a total score of 20 or higher on the EAT-26 as an exclusion criteria for data analysis. Cronbach’s α of the EAT-26 was 0.82.

International Physical Activity Questionnaire—Short Form

International Physical Activity Questionnaire—Short Form (IPAQ) is a seven-item self-report measure that assesses levels of physical activity (lasting for a minimum of 10 minutes) in the last 7 days (Craig et al., 2003). The IPAQ measures the frequency and duration of vigorous and moderate physical activity and time spent walking and sitting. The continuous scoring includes various formulae that are available for metabolic equivalent of task (MET)-minutes per week calculations for the various types of physical activity (Guidelines for Data Processing and Analysis of the IPAQ, 2005). The IPAQ has very good test–retest reliability (Spearman’s p of 0.76) (Craig et al., 2003).

Yale Food Addiction Scale

The Yale Food Addiction Scale (YFAS) is a 25-item self-report measure that assesses individuals’ addictive behaviors toward high-fat and high-sugar foods (Gearhardt et al., 2009). Participants were asked about the occurrence of different eating behaviors in the past 12 months, comparable to questions related to the diagnostic criteria for substance dependence. The YFAS uses 5-point Likert responses and has eight subscales that each have a cutoff point that designates whether the respondent meets criteria for that symptom of addiction (coded 0 = no, 1 = yes). The YFAS total score, with a possible range of 0–8, is the sum of these dichotomous scores and is what was used in this study. The YFAS has demonstrated good internal reliability (α = 0.86) and good convergent and discriminant validity (Gearhardt et al., 2009).

Statistical analysis

Data were assessed for normality. Data were non-normal if skew was 2 or higher and kurtosis was 7 or higher (West et al., 1995). A log transformation was successfully used to correct for non-normal skewness for negative affect at Time 1 and Time 2.

Consistent with previous research using the AAT (Rinck and Becker, 2007; Wiers et al., 2009, 2010), medians were used as variables to summarize each participant’s performance on the D-AAT. The use of medians (as opposed to means) is preferred because they reduce the influence of outliers and avoid the creation of arbitrary cutoffs to identify outliers. Data from error trials and trials occurring immediately after an error (i.e. trials that were repeated because of the error) were eliminated from analyses. Each participant had four median reaction times that were calculated for pulling and pushing both dessert and neutral images. Each participant had an “ApproachTend1” (approach tendency for desserts at time 1, calculated by (RT_{DessertPushT1} − RT_{DessertPullT1}) − (RT_{NeutralPushT1} − RT_{NeutralPullT1})) and “ApproachTend2” (approach tendency for desserts at time 2; calculated by (RT_{DessertPushT2} − RT_{DessertPullT2}) − (RT_{NeutralPushT2} − RT_{NeutralPullT2})). Subtracting ApproachTend1 from ApproachTend2 yielded a change score in approach tendency for dessert images, “Delta Approach.” Increases in approach bias to desserts were reflected by positive numbers on Delta Approach, whereas increases in withdrawal bias to dessert images were reflected by negative numbers.

A series of independent samples t-tests were conducted to determine differences between the experimental groups on various measures. An analysis of covariance (ANCOVA) was conducted with a main factor of Group 2 (Exercise vs Cognitive) and “Delta Approach” (ApproachTend2 − ApproachTend1) as the dependent variable, and any variables that were significantly different between the experimental groups were included as covariates.

Next, a one-tailed Pearson correlation was conducted to determine whether physical activity levels, as measured by the IPAQ (MET-minutes/week), was negatively related to baseline implicit approach motivation to dessert images (ApproachTend1).

Results

Characteristics of the study participants are summarized in Table 1 by condition (exercise vs cognitive task). There were no group differences on demographics or baseline hunger levels, but group differences emerged on the EAT-26 total score with individuals in the exercise condition reporting more disordered eating (mean = 5.69, SD = 4.55) than individuals in the cognitive condition (mean = 3.73, SD = 3.69), t(1, 86) = −2.15, p = 0.03 (Table 1). A two-tailed, independent samples t-test revealed that individuals in the cognitive condition (mean = 5.14, SD = 1.36) rated the cognitive task as being more difficult than the individuals in the exercise condition (mean = 3.50, SD = 1.74) rated the exercise task, t(1, 86) = 4.77, p < 0.001. Finally, individuals in the cognitive task condition had significantly greater increases in negative affect (Negative affect at Time 2 − Negative affect at Time 1; mean = 0.02, SD = 0.08) compared to individuals in the exercise condition (mean = −0.03, SD = 0.06), t(1, 86) = 3.07, p = 0.003. Due to group differences, EAT-26 total score, task difficulty, and changes in negative affect were used as covariates where appropriate in subsequent analyses.

Table 1.

Baseline characteristics of study population by condition.

	Exercise (n=51)	Cognitive (n=37)	p value
Variable
Age
Mean (SD)	18.78 (1.01)	18.81 (0.91)	0.90
Gender
Female (%)	56.90	56.80	0.99
Male (%)	43.10	43.20
Race
White (%)	62.70	51.40	0.11
Asian (%)	27.50	40.50
Black (%)	2.00	8.10
Other (%)	7.80	0.00
Ethnicity
Non-Hispanic (%)	92.20	97.20	0.32
Hispanic (%)	7.80	2.80
BMI	23.03 (3.50)	22.54 (2.73)	0.48
Body fat percentage	20.39 (6.94)	19.37 (6.63)	0.49
Weight control	4.50 (2.01)	4.00 (1.84)	0.40
Baseline physical activity (MET-minutes/week)	3707.66 (2288.69)	3254.73 (1967.36)	0.36
Minutes since last meal	320.86 (280.86)	322.03 (287.95)	0.99
Hunger level (SLIM)	4.67 (1.50)	4.64 (1.16)	0.90
Eating attitudes test	5.69 (4.55)	3.73 (3.69)	0.03
Food addiction (YFAS)	1.84 (1.58)	1.81 (1.41)	0.92
Positive affect Time 1 (PANAS)	21.59 (6.78)	23.16 (7.41)	0.30
Negative affect Time 1 (PANAS)^a	1.08 (0.09)	1.08 (0.07)	0.98
Positive affect Time 2 (PANAS)	22.96 (7.94)	22.78 (6.53)	0.91
Negative affect Time 2 (PANAS)	1.05 (0.07)	1.10 (0.09)	0.006
Change in PA (Time 2 − Time 1)	1.37 (4.85)	−0.38 (4.49)	0.09
Change in NA (Time 2 − Time 1)	−0.03 (0.06)	0.02 (0.08)	0.003
Task difficulty	3.50 (1.74)	5.14 (1.36)	<0.001

SD: standard deviation; BMI: body mass index; MET: metabolic equivalent of task; SLIM: Satiety Labeled Intensity Magnitude; YEAS: Yale Food Addiction Scale; PANAS: Positive and Negative Affect Schedule; PA: positive affect; NA: negative affect.

Negative affect Time 1 and Time 2 are log-transformed.

In total, participants were on average 18.8 years (SD = 0.9 years) and most were women (56.8%). In terms of race, 58.0 percent identified themselves as White, 33.0 percent as Asian, 4.5 percent as Black/African American, and 4.5 percent as Other. In terms of ethnicity, 5.7 percent were Hispanic/Latino, 93.2 percent were not Hispanic/Latino, and 1 did not report. Of the 88 participants, 51 (58.0%) were in the exercise condition and 37 (42.0%) were in the cognitive condition.

Group differences by experimental condition

See Table 2 for ApproachTend scores by condition and timing session. Groups did not differ with regard to Time 1 performance on the D-AAT (p = 0.31) or Time 2 performance on the D-AAT (p = 0.64).

Table 2.

ApproachTend scores by condition and time session.

	N	Mean (in ms)	SD
ApproachTend Time 1: exercise condition
Neutral images	51	−28.10	66.24
Dessert images	51	−35.83	70.15
ApproachTend1	51	−7.74	69.67
ApproachTend Time 1: cognitive condition
Neutral images	37	−15.46	49.78
Dessert images	37	−7.10	81.98
ApproachTend1	37	8.37	76.79
ApproachTend Time 2: exercise condition
Neutral images	51	−21.86	73.14
Dessert images	51	−17.57	60.15
ApproachTend2	51	4.29	69.91
ApproachTend Time 2: cognitive condition
Neutral images	37	−17.50	66.16
Dessert images	37	−19.85	63.05
ApproachTend2	37	−2.35	60.12

SD: standard deviation; ApproachTend: approach tendency.

Contrary to our hypothesis, the exercise group (M = 12.03, SD = 98.13) became significantly more approach oriented to the dessert images compared to the cognitive group (M = −10.72, SD = 101.55) after controlling for the effects of disordered eating (EAT-26), task difficulty, and changes in negative affect, F(1, 83) = 2.91, p = 0.046 (see Figure 1).

Figure 1.

Condition by timing session interaction using the exercise and cognitive conditions with approach tendency to dessert images as the dependent variable (Hypothesis 1).

Finally, there was no significant correlation between levels of physical activity and withdraw bias toward dessert images at Time 1, r(86) = 0.13, p = 0.12, but the data trended in the opposite direction of what was predicted.

Discussion

We found that college students randomized to the acute aerobic exercise condition became more approach oriented toward desserts compared to individuals in the cognitive task condition after controlling for group differences in disordered eating, task difficulty, and changes in negative affect. We also found that individuals with higher levels of physical activity tended to be more implicitly motivated for dessert food images compared to neutral images, although the results were not statistically significant.

Previous research has shown that short 5-minute bouts of exercise can decrease urges to snack (Thayer et al., 1993). In this study, it was expected that a 20-minute bout of exercise would also decrease implicit motivations to consume food, especially desserts. Contrary to our hypothesis, our findings indicated that there was a positive relationship between an acute bout of physical activity and approach bias for desserts. After exercise, participants were more approach oriented toward the dessert images relative to the neutral images.

Other research has shown that some individuals may see unhealthy foods (e.g. desserts) as more desirable after a workout (Finlayson et al., 2011). There are many factors that affect whether or not individuals find unhealthy desserts appetitive after exercise, including personality differences, personal exercise motivations, and weight loss or maintenance goals (Lowry et al., 2000; Ryan et al., 1997). Many of these factors are subject to explicit attitudes and motivations and do not take into account implicit motivations. The data from this study suggest that there is an implicit, “bottom-up” desire for food after a bout of exercise. From an evolutionary perspective, individuals need to consume energy (i.e. calories) in order to survive and maintain bodily functions. Woo and Pi-Sunyer (1985) found that lean women partook in a compensatory intake response after mild and moderate exercise. That is, active non-obese women ate more calories after exercise to compensate for the energy deficit from the exercise. However, this is not true for obese women. Other research has shown that active obese women did not increase their caloric intake after exercise (Woo et al., 1982). Because the sample in this study was mainly composed of average-weight undergraduates (only 13.6% participants were overweight and 4.5% were obese according to their body mass index (BMI)), it is possible that these individuals had a natural inclination to replenish the calories they lost during exercise. Normal weight undergraduates who exercise expend more energy and may, therefore, be more approach oriented to the food images because there is an evolutionary drive to replenish energy that has been expended.

All participants were college students at a private university who, when compared with their non-college counterparts, may have been healthier and less sedentary (Pampel et al., 2010). Therefore, the results may not be applicable to the general population in this age group. In the D-AAT used in this study, participants were only shown images of dessert foods as appetitive stimuli. Some participants may not find dessert foods appetitive and may prefer other “addicting” foods that fulfill the salty/crunchy cravings, such as potato chips. Previous research has shown that some of the more desirable comfort foods include potato chips, ice cream, cookies, and candy/chocolate (Wansink et al., 2003). Therefore, the reaction times that measured approach motivation for food in this study may be skewed to only reflect those individuals who find dessert foods extremely appetizing and may miss the individuals who find salty foods (e.g. potato chips) most desirable. As a result, future studies should include images of other salty and/or savory foods in addition to the dessert images to fully capture the individual differences in food preferences.

Future directions

In order to better understand how exercise affects implicit approach motivation to dessert foods, future studies could alter the experimental strategy used. Exercise intensity and duration could be altered in future studies. This study used a moderate-intensity bout of exercise, which did not appear to be difficult for the majority of participants in this sample. Previous research has shown that aerobic exercise intensity plays a role in short-term levels of hunger and food consumption (King et al., 1994; Laan et al., 2010). Increasing the intensity of the exercise may decrease the “bottom-up” approach motivation for foods. The lower intensity moderate exercise used in this study, however, may not have affected individuals’ appetitive drive toward food to the extent needed to produce a decrease in relative approach motivation for dessert food images. The types of images included in the D-AAT could be amended to include both sweet and savory foods, such as potato chips, to compare differences in implicit motivation to this different food type.

Also, it would be beneficial to create a training paradigm using the D-AAT. This training paradigm, which would include individuals implicitly learning to “avoid” dessert images, may help to alter the attentional biases that individuals have toward appetizing desserts. This type of training paradigm has been used in previous research with undergraduates who were identified as hazardous drinkers (Wiers et al., 2009) and with a clinical alcoholic population (Wiers et al., 2011). These studies using the alcohol-AAT show promising evidence that implicit bias can be altered through a training version of the AAT.

Conclusion

This study provides the first evidence that an acute bout of moderate-intensity aerobic exercise increases implicit relative approach motivation for dessert food images in young collegiate adults.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Bradley

Lang

(1994) Measuring emotion: The self-assessment manikin and the semantic differential. Journal of Behavioral Therapy and Experimental Psychiatry 25: 49–59.

Cacioppo

Priester

Berntson

(1993) Rudimentary determinants of attitudes. II: Arm flexion and extension have differential effects on attitudes. Journal of Personality and Social Psychology 65: 5–17.

Canadian Society for Exercise Physiology (2002) Physical activity readiness questionnaire (PAR-Q). Available at: http://www.csep.ca/english/view.asp?x=698

Cardello

Schutz

Lesher

et al . (2005) Development and testing of a labeled magnitude scale of perceived satiety. Appetite 44: 1–13.

Center for the Study of Emotion and Attention (1999) The International Affective Picture System: Digitized Photographs. Gainesville, FL: The Center for Research in Psychophysiology, University of Florida.

Centers for Disease Control and Prevention (CDC) (2011) Defining overweight and obesity. Available at: http://www.cdc.gov/obesity/adult/defining.html

Colic Baric

Satalic

Lukesic

(2003) Nutritive value of meals, dietary habits and nutritive status in Croatian university students according to gender. International Journal of Food Sciences and Nutrition 54: 473–484.

Craig

Marshall

Sjostrom

et al . (2003) International physical activity questionnaire: 12-country reliability and validity. Medicine & Science in Sports & Exercise 35: 1381–1395.

Daneman

Carpenter

(1980) Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior 19: 450–466.

10.

Dinger

Brittain

Hutchinson

(2014) Associations between physical activity and health-related factors in a national sample of college students. Journal of American College Health 62: 67–74.

11.

Elder

Roberts

(2007) The effects of exercise on food intake and body fatness: A summary of published studies. Nutrition Reviews 65: 1–19.

12.

Evero

Hackett

Clark

et al . (2012) Aerobic exercise reduces neuronal responses in food reward brain regions. Journal of Applied Physiology 112: 1612–1619.

13.

Faul

Erdfelder

Buchner

et al . (2009) Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods 41: 1149–1160.

14.

Finlayson

Caudwell

Gibbons

et al . (2011) Low fat loss response after medium-term supervised exercise in obese is associated with exercise-induced increase in food reward. Journal of Obesity. Epub ahead of print 20 September. DOI: 10.1155/2011/615624

15.

Fox

3rd Naughton

Haskell

(1971) Physical activity and the prevention of coronary heart disease. Annals of Clinical Research 3: 404–432.

16.

Gable

Harmon-Jones

(2008) Approach-motivated positive affect reduces breadth of attention: Research article. Psychological Science 19: 476–482.

17.

Garner

Olmsted

Bohr

et al . (1982) The eating attitudes test: Psychometric features and clinical correlates. Psychological Medicine 12: 871–878.

18.

Gearhardt

Corbin

Brownell

(2009) Preliminary validation of the Yale Food Addiction Scale. Appetite 52: 430–436.

19.

Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ)—Short and Long Forms (2005). IPAQ Web site. http://www.ipaq.ki.se/dloads/IPAQ%20LS%20Scoring%20Protocols_Nov05.pdf

20.

Juergensen

Demaree

(2015) Approach-motivated positive affect and emotion regulation alter global–local focus and food choice. Motivation and Emotion 39: 580–588.

21.

King

(1991) The natural history of eating pathology in attenders to primary medical care. International Journal of Eating Disorders 10: 379–387.

22.

King

Burley

Blundell

(1994) Exercise-induced suppression of appetite: Effects on food intake and implications for energy balance. Eur J Clin Nutr 48: 715–724.

23.

Krieglmeyer

De Houwer

Deutsch

(2013) On the nature of automatically triggered approach-avoidance behavior. Emotion Review 5: 280–284.

24.

Laan

Leidy

Lim

et al . (2010) Effects and reproducibility of aerobic and resistance exercise on appetite and energy intake in young, physically active adults. Applied Physiology, Nutrition, and Metabolism 35: 842–847.

25.

Lowry

Galuska

Fulton

et al . (2000) Physical activity, food choice, and weight management goals and practices among US college students. American Journal of Preventive Medicine 18: 18–27.

26.

May

Juergensen

Demaree

(2016) Yum, cake!: How reward sensitivity relates to automatic approach motivation for dessert food images. Personality and Individual Differences 90: 265–268.

27.

Taylor

(2012) Brisk walking reduces ad libitum snacking in regular chocolate eaters during a workplace simulation. Appetite 58: 387–392.

28.

Pampel

Krueger

Denney

(2010) Socioeconomic disparities in health behaviors. Annual Review of Sociology 36: 349–370.

29.

Rinck

Becker

(2007) Approach and avoidance in fear of spiders. Journal of Behavior Therapy and Experimental Psychiatry 38: 105–120.

30.

Ryan

Frederick

Lepes

et al . (1997) Intrinsic motivation and exercise adherence. International Journal of Sport Psychology 28: 335–354.

31.

Schienle

Schafer

Hermann

et al . (2009) Binge-eating disorder: Reward sensitivity and brain activation to images of food. Biological Psychiatry 65: 654–661.

32.

Taylor

Oliver

(2009) Acute effects of brisk walking on urges to eat chocolate, affect, and responses to a stressor and chocolate cue. An experimental study. Appetite 52: 155–160.

33.

Thayer

Peters

III Takahashi

et al . (1993) Mood and behavior (smoking and sugar snacking) following moderate exercise: A partial test of self-regulation theory. Personality and Individual Differences 14: 97–104.

34.

Turner

Engle

(1989) Is working memory capacity task dependent? Journal of Memory and Language 28: 127–154.

35.

Wansink

Cheney

Chan

(2003) Exploring comfort food preferences across age and gender. Physiology & Behavior 79: 739–747.

36.

Watson

Clark

Tellegen

(1988) Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 54: 1063–1070.

37.

West

Finch

Curran

(1995) Structural Equation Models with Nonnormal Variables: Problems and Remedies. Thousand Oaks, CA: SAGE.

38.

Wiers

Eberl

Rinck

et al . (2011) Retraining automatic action tendencies changes alcoholic patients’ approach bias for alcohol and improves treatment outcome. Psychological Science 22: 490–497.

39.

Wiers

Rinck

Dictus

et al . (2009) Relatively strong automatic appetitive action-tendencies in male carriers of the OPRM1 G-allele. Genes, Brain and Behavior 8: 101–106.

40.

Wiers

Rinck

Kordts

et al . (2010) Retraining automatic action-tendencies to approach alcohol in hazardous drinkers. Addiction 105: 279–287.

41.

Woo

Garrow

Pi-Sunyer

(1982) Effect of exercise on spontaneous calorie intake in obesity. American Journal of Clinical Nutrition 36: 470–477.

42.

Woo

Pi-Sunyer

. (1985) Effect of increased physical activity on voluntary intake in lean women. Metabolism 34: 836–841.

43.

World Health Organization (2003) Global Strategies on diet, physical activity and health—Obesity and overweight. Available at: http://www.who.int/hpr/NPH/docs/gs_obesity.pdf