Applied Research method

lable at ScienceDirect

Appetite 100 (2016) 210e215

Contents lists avai

Appetite

journal homepage: www.elsevier.com/locate/appet

The effect of academic stress and attachment stress on stress-eaters and stress-undereaters

Michael Emond a, *, Kayla Ten Eycke b, Stacey Kosmerly c, Adele Lafrance Robinson a, Amanda Stillar d, Sherry Van Blyderveen e

a Laurentian University, Sudbury, ON, Canada b University of Victoria, Victoria, BC, Canada c University of Ottawa, Ottawa, ON, Canada d University of Alberta, Edmonton, AB, Canada e McMaster University, Hamilton, ON, Canada

a r t i c l e i n f o

Article history: Received 17 August 2015 Received in revised form 27 January 2016 Accepted 30 January 2016 Available online 2 February 2016

Keywords: Stress-eating Eating behavior Attachment Stress

* Corresponding author. Department of Psycholog Ramsey Lk Rd., Sudbury, ON, P3E 2C6, Canada.

E-mail address: [email protected] (M. Emon

http://dx.doi.org/10.1016/j.appet.2016.01.035 0195-6663/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

It is well established that stress is related to changes in eating patterns. Some individuals are more likely to increase their overall food intake under conditions of stress, whereas others are more likely to consume less food when stressed. Attachment style has been linked to disordered eating and eating disorders; however, comparisons of eating behaviors under attachment versus other types of stress have yet to be explored. The present laboratory study examined the eating patterns in self-identified stress- undereaters and stress-eaters under various types of stress. More specifically, the study examined the effects of academic and attachment stress on calorie, carbohydrate and sugar consumption within these two groups. Under the guise of critiquing student films, university students viewed either one of two stress-inducing videos (academic stress or attachment stress, both designed to be emotionally arousing) or a control video (designed to be emotionally neutral), and their food intake was recorded. Results demonstrated that the video manipulations were effective in inducing stress. Differential patterns of eating were noted based on group and stress condition. Specifically, stress-undereaters ate fewer calo- ries, carbohydrates and sugars than stress-eaters in the academic stress condition, but not in the attachment stress or control condition. Findings suggest that specific types of stressors may influence eating behaviors differently.

© 2016 Elsevier Ltd. All rights reserved.

It is well established that a relationship exists between stress and changes in both food choice and food intake (Adam & Epel, 2007; Greeno & Wing, 1994; Torres & Nowson, 2007). In terms of food choice, the literature suggests that individuals demonstrate a shift toward more “unhealthy” food options when stressed. In particular, people tend to increase their intake of highly palatable, snack-type foods when stressed (O'Connor, Jones, Conner, McMillan, & Ferguson, 2008; Oliver & Wardle, 1999; Wansink, Cheney, & Chan, 2003; Zellner et al., 2006). For example, stress has been associated with increased consumption of sweet (Kandiah, Yake, Meyer, & Jones, 2006), salty (Wardle, Steptoe, Oliver, & Lipsey, 2000) and high fat food options (Kandiah et al.,

y, Laurentian University, 935

d).

2006; Wansink et al., 2003), whereas individuals tend to report a decrease in their consumption of more meal-type foods (i.e., meats, fruits, vegetables) under conditions of stress (Oliver & Wardle, 1999; Wansink et al., 2003; Zellner et al., 2006).

Research has also linked stress to changes in food intake. Many studies have demonstrated that stress has a bidirectional effect on food intake, whereby some individuals increase and others decrease their food consumption when stressed (Epel et al., 2004; Kandiah et al., 2006; Kandiah, Yake, & Willett, 2008; Oliver & Wardle, 1999; Wallis & Hetherington, 2009). Some studies have found that almost equal proportions of individuals increase versus decrease their food intake under stress, with approximately half of the sample reporting eating less and approximately half of the sample eating more under stress than they would otherwise (Oliver & Wardle, 1999; Stone & Brownell, 1994; Wallis & Hetherington, 2009). Other studies, however, have reported different distribu- tions of food intake changes under stress, with some finding

M. Emond et al. / Appetite 100 (2016) 210e215 211

roughly twice as many participants increased their food intake relative to those who decreased their food intake (Kandiah et al., 2006; 2008). These differences in findings may reflect differences in the sample populations or type of stressor.

Previous research has considered eating in response to a variety of stressors, yet few studies have considered whether individuals' eating responses vary depending on the stressor. Stress is defined in the literature as a disturbance to one's “physiological homeostasis or psychological well-being” (National Research Council, 2008, p. 2). Within the literature stress has been defined, differentiated and manipulated in a wide variety of ways. Common methods for inducing stress in a laboratory setting include inducing a high cognitive load on the participants (Lattimore, 2001; Wallis & Hetherington, 2004, 2009), or having the participants recall a recent stressful event (Sinha, Catapano, & O'Malley, 1999). Stress has also been defined to include mood manipulation through viewing a mildly distressing video (Grunberg & Straub, 1992) and through physical threat, such as fear of an impending shock (Heatherton, Herman, & Polivy, 1991; Herman & Polivy, 1975). The present study focused on two stressors: attachment stress, that is emotional stress that arises from anxiety related to an individual's relationship with their primary caregiver(s), and academic stress, that is a psychological stress that arises from the anticipation of evaluation, and in this case, in the context of academics. Videos related to each of these types of stress were created for the purpose of the present study in order to induce stress in participants. In addition, an emotionally neutral video about travel was also created, and used as a control.

Attachment refers to an internal schema of relating to the world, which is developed through repeated early interactions with pri- mary caregivers (e.g., mother) (Bowlby, 1980). Securely attached individuals internalize a sense of safety and trust in others, which allows them to self-soothe in times of distress (van den Boom, 1994). Since attachment influences one's perception of and in- teractions with their environment (Shorey & Snyder, 2006), it is likely an important factor to consider in relation to the stress-eating relationship.

Maunder and Hunter (2001) proposed an attachment model developed from a biopsychosocial perspective to explain how the relationship between a child and their caregiver can later affect the child's stress-response in adulthood (Maunder & Hunter, 2001). Their model describes how attachment insecurity can lead to increased risk for illness through increased vulnerability to stress, increased reliance on external affect regulation, and alternative methods of seeking help. In support of this view, animal and human studies alike have demonstrated that attachment is a key factor involved in the physiological response to stress (i.e., Lui et al., 1997; Powers, Pietromonaco, Gunlicks, & Sayer, 2006; Spangler & Schieche, 1998), eating behavior, and food choice (Koskina & Giovazolias, 2010; Troisi & Gabriel, 2014; Wilkinson, Rowe, Bishop, & Brunsrom, 2010). In fact, a laboratory study of eating behavior observed that individuals who received an anxious attachment prime ate significantly more than individuals who received a secure attachment prime (Wilkinson, Rowe, & Heath, 2013).

Academic stress (or test anxiety) refers to the stress associated with the academic environment, writing tests, performing difficult cognitive tasks, or being evaluated (Beggs, Shields, & Janiszewski Goodin, 2011). This stressor was utilized in the current study due to its relevance in undergraduate samples (Ross, Niebling, & Heckert, 1999) and the fact that it has been shown to be an effec- tive stressor in studies examining stress and eating. For instance, Pollard, Steptoe, Canaan, Davies, and Wardle (1995) compared a control group (not taking exams) to an exam-stress group just prior to their examination period. Stress measures and dietary

assessments were made in both groups during a baseline period to provide within-group comparisons. The exam-stress group re- ported significant increases in perceived stress at the exam sessions compared to their baseline sessions and participants with high trait anxiety showed significant increases in their caloric intake. In a later study, using a within-subject design, Costarelli and Patsai (2012), showed that female students reported significantly higher anxiety and disordered eating attitudes during examination compared to a control period. Furthermore, Feld and Shusterman (2015) found that students experience both increased physical and psychological stress and a change in their eating habits while in an academic environment.

1. Objectives

The present study sought to examine the impact of two types of stressors on food intake and food choice in a controlled experi- mental setting. The primary objective of the study was to compare the ways in which two distinct stressors, academic stress and attachment stress, might influence eating behaviors in two stress- eating populations: stress-undereaters and stress-eaters. Although it was predicted that stress-eaters would increase their intake in the face of stress, while stress-undereaters would decrease their intake, the present study was exploratory in exam- ining the ways in which the two distinct stressors, academic stress and attachment stress, might influence eating behaviors in these stress-eating populations.

2. Materials & methods

2.1. Participants

Participants included 167 female university students between the ages of 17 and 42 (M ¼ 20 years old, SD ¼ 3.21) who self- identified as being either a stress-undereater or a stress-eater. Those who reported “eating the same” under stress were excluded from the analysis (14.1%, n ¼ 29). Participants were excluded from the present study if they: reported a history of an eating disorder (2.4%, n ¼ 4), were pregnant or nursing (1.2%, n ¼ 2), reported being “very full” or “full” upon arrival to the study session (12.4%, n ¼ 20), or if they guessed the true nature of the study before it was revealed to them (17.7%, n ¼ 25). T-tests revealed that these participants excluded for guessing the true nature of the study did not differ from the remainder of the sample in terms of nutrient consumption or state anxiety score. In total, 51 partici- pants (30%) from the original sample were excluded from the present analyses, leaving 116 usable cases. All participants provided written consent to participate after they were debriefed about the true nature of the study.

2.2. Measures

2.2.1. Body measurements Participants' waist-to-hip ratios (WHR) were calculated, as

described by Klein et al., (2007). This was done with a flexible measuring tape. Participants’ BMIs were calculated by measuring their weight in kilograms and dividing it by their height in meters, squared. This was done with a flexible measuring tape and a digital bathroom scale. On average, participants weighed 65.81 kg (SD ¼ 13.22) and were 166.61 cm tall (SD ¼ 7.16). The average body mass index (BMI) was 24.11 (SD ¼ 4.89).

2.2.2. State-trait anxiety inventory (STAI; Spielberger, Gorsuch, & Lushene, 1970)

The STAI is a self-report measure comprised of two subscales

M. Emond et al. / Appetite 100 (2016) 210e215212

reflecting current, temporary “state” anxiety, and long-standing “trait” anxiety. This scale is comprised of 40 items (20 that measure state anxiety and 20 that measure trait anxiety). For each item, participants rate a statement on a 4-point Likert scale ranging from “Almost Never” to “Almost Always”. Possible scores range from 20 to 80, with higher scores indicating greater anxiety. This scale is appropriate for individuals with at least a sixth grade reading level. The STAI is a robust and widely used measure of stress and anxiety. In fact, it is possibly the most long-standing and frequently used measure of stress (Bieling, Antony, & Swinson, 1998). The STAI has good psychometric properties (internal consistency: .86 to .95; test-retest reliability over 2 months: .65 to .75; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983). Spielberger (1989) has also demonstrated that this scale has good construct and concur- rent validity. For the current sample, the Cronbach's alpha for this scale was .56.1 The average Trait Anxiety Inventory score for the sample was 40.78 (SD ¼ 9.61).

2.2.3. Stress-eating and stress-undereating Participants self-identified as “stress-eaters” or “stress-under-

eaters”. Participants were asked, “In general, which describes your eating behavior under stress compared to a relaxed state?” Par- ticipants responded using the following Likert scale options: undereat, slightly undereat, eat the same, slightly overeat, and overeat. Participants were categorized, based on their responses, into two groups: stress-eaters (slightly overeat or overeat) and stress-undereaters (slightly undereat, undereat), those who re- ported “eating the same” were excluded from the analysis.

2.3. Procedure

Data were collected from two institutions. All research assis- tants were trained by one lead researcher in order to ensure con- sistency in testing procedures across study sites. Participant stress was manipulated and behavioral eating responses were observed. Participants were recruited under the guise of critiquing videos for a student film competition. Each participant attended a film viewing session individually, and their stress levels were manipu- lated by exposing them to one of three 30 min videos: one designed to elicit attachment stress, one designed to elicit academic stress, and one neutral/relaxing video. While viewing the film, partici- pants had access to a wide selection of foods covering a range of palatability and nutritional content. The methodology used in this study parallels the previous research of Oliver and Wardle (1999), Patel and Schlundt (2001), Cartwright et al. (2003), Zellner et al. (2006) and Kandiah et al. (2006). This study was approved by the Research Ethics Board at the institutions where this study was conducted.

Experimental sessions took place in a small classroom or office with a table or desk, a computer and a chair. Testing occurred anywhere between 9:30 am and 6 pm. The experimenter informed participants they would be watching a film made by local students, which they would evaluate afterwards. This guise was used to prevent participants from being aware of the true nature of the experiment and potentially adjusting their eating behavior (i.e., participant expectancy effects). Videos (see description below) were presented on a computer and lights were turned off during the video.

During each session, a wide variety of foods (see food

1 It should be noted that this is an unexpectedly low value of the Cronbach alpha for the STAI. It is not clear why this particular sample would produce this low value. Following Sijtsma's (2009) argument about the limited usefulness of this statistic, the decision was made to include the STAI despite this low value.

descriptions below) were set out on the table near the participant prior to the experimenter leaving the room for the remaining duration of the video. Participants were told that these foods were “leftover food from a faculty meeting” and were a “thank you for their participation”. All of the food had been weighed on a digital OXO™ kitchen food scale prior to the participant's arrival.

Following the video, participants were asked to evaluate the film. Next, participants completed the state portion of the STAI. This measure served as a manipulation check, in that it measured whether participants in the two stress conditions reported expe- riencing more stress than participants in the control condition. Participants were subsequenlty debriefed about the true nature of the study and written consent for the use of their data was ob- tained. The posteexperiment questionnaire was then adminis- tered. Participants were also asked to report whether they were aware of the true intentions of the study. Finally, participant waist, hip, height, and weight measurements were taken and, once the participant left, the remaining food was weighed to determine the amount of each food eaten during the session.

2.3.1. Video manipulation Three 30-min videos were produced by the experimenters for

the experimental manipulation. The videos contained photo still frames, interviews, facts, music and short dramatizations; and were focused on one of three topics: travel (control group), exam stress (academic stress group), or maternal relationships (attachment stress group). The travel video was designed to elicit either a neutral or relaxed state in participants, while the academic stress and attachment relationship videos were designed to elicit stressful reactions. Although the current videos were novel, the use of videos as a stressor has been done in previous research and has been shown to successfully elicit stress in participants, independent of participant gender. Verification in past studies has included STAI, cortisol levels and heart rate (for example, the use of a video on industrial accidents: Grunberg & Straub, 1992; the use of a video on corneal-transplant surgeries: Takai et al., 2004; the use of a video on woodshop safety: Gilbert, Robinson, Chamberlin, & Spielberger, 1989).

2.3.2. Food selection The following foods and beverages were provided in clear

plastic serving trays or bowls: strawberries, sliced oranges, red grapes, pretzels, plain M&M™ candies, plain Tostitos™ tortilla chips, Doritos™, President's Choice™ mild salsa, Runtz™ hard candies, gummy bears, Cherry Tomatoes, Vegetable Thins™, Babybell cheese™ and President's Choice™ bottled water. Approximately three recommended servings (suggested by manufacturer or by nutritiondata.com) of each food item were provided. Nutrition information was retrieved from food packaging, or from nutritiondata.com for produce. These foods were chosen based on methodological feasibility, past research, and the oppor- tunity to offer a range of flavors and nutritional content.

2.4. Data analysis

The data were first analyzed using a one-way analysis of vari- ance (ANOVA) to determine whether the academic and attachment videos did in fact induce a state of stress in the participants. The data was then analyzed using a between-subjects multivariate analysis of variance (MANOVA) to examine possible differences in food consumption based on stress-eating style and stress condition. The three dependent variables were total calories consumed, total carbohydrates consumed and total sugar consumed. The two in- dependent variables were stress-eating style (stress-undereater, stress-eater) and stress condition (attachment stress, academic

Table 1 Means, standard errors and p-values for calories consumed by stress-undereaters and stress-eaters under conditions: academic stress, attachment stress, control.

Condition Stress-eating style M SE N p

Control Stress-undereater 120 70.0 12 .81 Stress-eater 135 45.8 28

Attachment stress Stress-undereater 326 62.6 15 .43 Stress-eater 405 57.2 18

Academic stress Stress-undereater 90 73.1 11 .04 Stress-eater 291 42.9 32

Table 2 Means, standard deviations and p-values for carbohydrates consumed by stress- undereaters and stress-eaters under conditions: control, attachment stress, aca- demic stress.

Condition Stress-eating style M SE N p

Control Stress-undereater 35 12.6 12 .99 Stress-eater 35 8.3 28

Attachment stress Stress-undereater 58 11.3 15 .52 Stress-eater 69 10.3 18

Academic stress Stress-undereater 18 13.2 11 .003 Stress-eater 65 7.7 32

M. Emond et al. / Appetite 100 (2016) 210e215 213

stress, control). Pillai's trace was the criterion used. Upon finding significant results, follow-up univariate ANOVAs were conducted on each dependent measure separately to determine the locus of the statistically significant multivariate interaction effect.

3. Results

3.1. Stress-induction

Participants were asked to rate the level of stress they experi- enced as a result of the video. A chi-square test was used to analyze whether the rated levels of stress differed after exposure to the three video manipulations. A significant main effect was found in stress ratings of the videos (c2 ¼ 27.42, df ¼ 6, p < .001) with the attachment and academic stress being significantly higher than the control video. The average score on the STAI state anxiety scale for the entire sample was 36.82 (SD ¼ 11.65). Further, a one-way ANOVA was run to determine whether there were significant dif- ferences in STAI scores between the stress conditions. A significant effect was found when comparing state STAI scores between attachment (M ¼ 33.81, SD ¼ 11.08), academic (M ¼ 46.08, SD ¼ 11.95), and control (M ¼ 31.40, SD ¼ 4.97) groups [F(2,79) ¼ 15.98, p < 0 .001]. Both the academic stress and the attachment stress conditions induced significantly higher STAI scores than the control condition. Additonally, the academic stress video induced significantly higher STAI scores than the attachment video.

3.2. Differences in food consumption based on stress-eating style and stress condition

A between-subjects MANOVA was performed on three depen- dent variables: total calories consumed, total carbohydrates consumed and total sugars consumed. The independent variables were stress-eating style (stress-undereater, stress-eater) and stress condition (attachment stress, academic stress, control). The multivariate interaction effect of stress-eating style � stress con- dition was statistically significant, Pillai's trace ¼ .24, F(15,330) ¼ 1.88, p ¼ .02, indicating that the unique joint effect of these two variables did account for a significant portion of the variance. The dependent variables: total calories consumed, total carbohydrates consumed, and total sugars consumed, were each significantly affected by the interaction effect of stress-eating style � stress condition, F(5,11) ¼ 4.8, p ¼ .001; F(5,110) ¼ 3.6, p ¼ .005; F(5, 110) ¼ 3.7, p ¼ .004, respectively. Follow-up ANOVAs were conducted on each dependent measure separately to deter- mine the locus of the statistically significant multivariate interac- tion effect.

3.3. Calories

A one-way ANOVA was used to explore the difference in total caloric intake between stress-undereaters and stress-eaters in each of the stress conditions. Stress-eaters ate significantly more calories than stress-undereaters in the academic stress condition, F(1,41) ¼ 4.7, p ¼ .04. However, there was no statistically significant difference between total calories consumed by stress-undereaters and stress-eaters in the attachment stress condition, F(1,31) ¼ .65, p ¼ .43 or in the control condition, F(1,38) ¼ .06, p ¼ .8. See Table 1.

3.4. Carbohydrates

A one-way ANOVA was used to explore the difference in total carbohydrate intake between stress-undereaters and stress-eaters

in each of the stress conditions. There was no statistically signifi- cant difference between total carbohydrates consumed by stress- undereaters and stress-eaters in the attachment stress condition, F(1,31) ¼ .42, p ¼ .52 or in the control condition, F(1,38) ¼ .0, p ¼ 1. Yet, stress-eaters ate significantly more carbohydrates than stress- undereaters in the academic stress condition, F(1,41) ¼ 9.6, p ¼ .003. See Table 2.

3.5. Sugar

A one-way ANOVA was used to explore the difference in total sugar intake between stress-undereaters and stress-eaters in each of the stress conditions. Stress-eaters ate significantly more sugar than stress-undereaters in the academic stress conditions, F(1,41) ¼ 8.7, p ¼ .005. However, there was no statistically signifi- cant difference between total sugar consumed by stress- undereaters and stress-eaters in the attachment stress condition, F(1,31) ¼ .75, p ¼ .39 or in the control condition, F(1,38) ¼ .48, p ¼ 49. See Table 3.

4. Discussion

The present study examined the differential effects of stress on eating in self-identified stress-eaters and stress-undereaters in the laboratory. Overall, the results revealed that stress-undereaters ate fewer calories, carbohydrates and sugars than did stress-eaters; however, this was only observed in the academic stress condition. That the two stress-eating groups did not differ in their food intake in the control or attachment stress conditions suggests that both stress overall and certain types of stress are influencing eating behaviors. It is an important distinction to highlight, for if stress- eaters were found to eat more than stress-undereaters in all con- ditions, our stress-eating classification would in actuality be an “over-eating classification” and therefore unrelated to the changes in eating brought on by specific stressors. Since the differences between groups only manifested themselves in the academic stress condition, it both validates our measure for characterizing partici- pants into these two groups and demonstrates a need to further explore this classification in future research. Prior studies exam- ining the effects of stress on eating did not make this distinction and, therefore, assumed that a stressor's influence on food intake

Table 3 Means, standard deviations and p-values for sugar consumed by stress-undereaters and stress-eaters under conditions: control, attachment stress and academic stress.

Condition Stress-eating style M SE N p

Control Stress-undereater 21 9.5 12 .50 Stress-eater 27 6.2 28

Attachment stress Stress-undereater 41 8.5 15 .39 Stress-eater 53 7.7 18

Academic stress Stress-undereater 13 9.9 11 .005 Stress-eater 47 5.8 32

M. Emond et al. / Appetite 100 (2016) 210e215214

was unidirectional, thus limiting the ability to determine the real effect of stress on eating behavior. For example, Oliver, Wardle, and Gibson (2000) induced stress in their participants by having them prepare a 4-min speech that they expected to have filmed and assessed. Participants could choose between sweet, salty or bland, and high- or low-fat foods. Stress did not alter their overall intake of food (Oliver et al., 2000). Although this study made the distinction of emotional and restrained eating, it is possible that non- significant results reported in that study were due to the fact that stress-eaters and stress-undereaters cannot be fully classified looking at those variables alone.

The present study also attempted to present a wider variety of food choice than past studies. For example, in an experimental study that showed women were more likely to eat sweet foods and bland foods during a stressful video about industrial accidents (Grunberg & Straub, 1992) their choice of foods was limited to M&M's™, peanuts and crackers. In another study, Zellner et al. (2006), using unsolvable anagrams as a stressor, presented fe- males with a limited variety of foods from which to select (M&M's™, potato chips, dry roasted peanuts, red grapes). The stressed group ate more M&M's™ and fewer grapes than the control group, suggesting an increased preference for the selection of sweet, high-calorie, “unhealthy” foods during stress (Zellner et al., 2006). In line with these previous studies, our findings also support the notion that stress causes an increase in sweet food (sugar) intake, at least in those individuals who self-identify as “stress-eaters” and only when experiencing academic stress.

In terms of total caloric intake, our results revealed that stress- eaters ate more overall calories than stress-undereaters when exposed to academic stress, but both groups of eaters elevated their caloric intake when faced with attachment stress. In other words, our results suggest that not all stressors lead to the same behavioral adaptations. It is possible that attachment stress activated a different emotional response in the participants, leading both types of eaters to increase their food intake. Numerous studies have found links between attachment difficulties and disordered eating in non-clinical samples (Elgin & Pritchard, 2006; Ty & Francis, 2013; Wilkinson et al., 2010). For example, Wilkinson et al. (2010) conducted a survey study and reported that attachment insecu- rity was a predictor of disinhibited eating. There is also ample research in the clinical domain of eating disorders linking attach- ment to disordered eating (Tasca & Balfour, 2014; Tasca et al., 2009; Ward, Ramsay & Treasure, 2000). Also noteworthy is that the present study limited itself to female participants because of past findings showing that women are more likely to alter their food consumption under stress when compared to their male counter- parts (Grunberg & Straub, 1992; O’Connor et al., 2008; Pollard et al., 1995; Weinstein, Shide, & Rolls, 1997; Zellner et al., 2006). How- ever, future studies should explore the possibility for gender dif- ferences in stress-eaters and stress-undereaters, especially in the face of attachment stress, since the effects of attachment difficulties on eating behaviors are not exclusive to women (Elgin & Pritchard, 2006).

The present study is the first of its kind to examine the effects of different stressors on food intake and preference in two distinct stress-eating populations. Our findings revealed group differences according to the stress-eating classification in that, when experi- encing academic stress, self-reported stress-eaters consumed more calories, carbohydrates and sugars than did stress-undereaters, therefore, lending support to the existence of this phenomenon in a real world setting.

This study is not without limitations. For example, the foods presented to participants did not allow for an analysis of fat intake, as few of the options were high-fat and animal fats were not made available. Even though this study included more food choices than previous research, future studies must include an even wider array of choices, in particular fats, in order to determine the impact of stress and stress-eating style on other macronutrients.

Ideally, baseline stress would have been measured in the study, however due to the deceptive nature of the manipulation it was felt that if the participants filled out a STAI before the viewing of the videos then the number of participants aware of the nature of the study would have increased. A further limitation of the study regarding the STAI was its low Cronbach's alpha value indicating low internal consistency. While this could be due to chance, it is important that additional measures of stress or emotional distress should be included in future studies to further verify the validity of the stress manipulation. Additionally, while this study did include a self-report measures of the participants' current eating patterns, future studies should include objective measures of dietary re- straint as this has been shown to be one potential factor in pro- ducing stress-eating responses (Freeman & Gil, 2004).

While it is not possible from this current study to fully explain the differential effects of the stressors on the stress-eating response, the pattern of results clearly suggests a need for the field to turn its attention to the influence of the type of stressor, given that participants responded very differently in the context of academic stress versus attachment stress.

References

Adam, T. C., & Epel, E. P. (2007). Stress eating and the reward system. Physiology & Behavior, 91, 449e458.

Beggs, C., Shields, D., & Janiszewski Goodin, H. (2011). Using guided reflection to reduce test anxiety in nursing students. Journal of Holistic Nursing, 29, 140e147.

Bieling, P. J., Antony, M. M., & Swinson, R. P. (1998). The stateetrait anxiety in- ventory, trait version: structure and content re-examined. Behaviour Research and Therapy, 36, 777e788.

van den Boom, D. (1994). The influence of temperament and mothering on attachment and exploration: an experimental manipulation of sensitive responsiveness among lower-class mothers with irritable infants. Child Devel- opment, 65(5), 1457e1477.

Bowlby, J. (1980). Attachment and loss. Vol. III. Loss, sadness and depression. London: The Hogarth Press and the Institute of Psycho-Analysis.

Cartwright, M., Wardle, J., Steggles, N., Simon, A. E., Croker, H. C., & Jarvis, M. J. (2003). Stress and dietary practices in adolescents. Health Psychology, 22, 362e369.

Costarelli, V., & Patsai, A. (2012). Academic examination stress increases disordered eating symptomatology in female university students. Eating and Weight Dis- orders, 17(3), e164ee169.

Elgin, J., & Pritchard, M. (2006). Adult attachment and disordered eating in un- dergraduate men and women. Journal of College Student Psychotherapy, 21(2), 25e40.

Epel, E., Jimenez, S., Brownell, K., Stroud, L., Stoney, C., & Niaura, R. (2004). Are stress eaters at risk for the metabolic syndrome. Annals of the New York Academy of Sciences, 1032, 208e210.

Feld, L. D., & Shusterman, A. (2015). Into the pressure cooker: student stress in college prepatory high schools. Journal of Adolescence, 41, 31e42.

Freeman, L., & Gil, K. (2004). Daily stress, coping, and dietary restraint in binge eating. International Journal of Eating Disorders, 36(2), 204e212.

Gilbert, D. G., Robinson, J. H., Chamberlin, C. L., & Spielberger, C. D. (1989). Effects of smoking/nicotine on anxiety, heart rate, and lateralization of EEG during a stressful movie. Psychophysiology, 26(3), 311e320.

Greeno, C. G., & Wing, R. R. (1994). Stress-induced eating. Psychological Bulletin, 115, 444e464.

Grunberg, N. E., & Straub, R. O. (1992). The role of gender and taste class in the

M. Emond et al. / Appetite 100 (2016) 210e215 215

effects of stress on eating. Health Psychology, 11(2), 97e100. Heatherton, T., Herman, P., & Polivy, J. (1991). Effects of physical threat and ego

threat on eating behaviour. Journal of Personality and Social Psychology, 60, 138e143.

Herman, C. P., & Polivy, J. (1975). Anxiety, restraint, and eating behaviour. Journal of Abnormal Psychology, 84(6), 666e672.

Kandiah, J., Yake, M., Meyer, M., & Jones, J. (2006). Stress influences appetite and comfort food preferences in college women. Nutrition Research, 26, 118e123.

Kandiah, J., Yake, M., & Willett, H. (2008). Effects of stress on eating practices among adults. Family and Consumer Sciences Research Journal, 37(1), 27e38.

Klein, S., Allison, D. B., Heymsfield, S. B., Kelley, D. E., Leibel, R. L., Nonas, C., et al. (2007). Waist circumference and cardiometabolic risk: a consensus statement from shaping America's health: association for weight management and obesity prevention; NAASO, the obesity society; the American society for nutrition; and the American diabetes association. American Journal of Clinical Nutrition, 85(5), 1197e1202.

Koskina, N., & Giovazolias, T. (2010). The effect of attachment insecurity in the development of eating disturbances across gender: the role of body dissatis- faction. The Journal of Psychology, 144(5), 449e471.

Lattimore, P. J. (2001). Stress-induced eating: an alternative method for inducing ego-threatening stress. Appetite, 36(2), 187e188.

Lui, D., Diorio, J., Tannenbaum, B., Caldij, C., Francis, D., Freedman, A., et al. (1997). Maternal care, hippocampal glucocorticoid receptors, and hypothalamic- pituitary-adrenal responses to stress. Science, 227(5332), 1659e1662.

Maunder, R., & Hunter, J. (2001). Attachment and psychosomatic medicine: devel- opmental contributions to stress and disease. American Psychosomatic Society, 63, 556e567.

National Research Council (US) Committee on Recognition and Alleviation of Distress in Laboratory Animals. (2008). Recognition and alleviation of distress in laboratory animals (p. 2). Washington (DC): National Academies Press.

Oliver, G., & Wardle, J. (1999). Perceived effects of stress on food choice. Physiology and Behaviour, 66(3), 511e515.

Oliver, G., Wardle, J., & Gibson, E. L. (2000). Stress and food choice: a laboratory study. Psychosomatic Medicine, 62(6), 853e865.

O'Connor, D. B., Jones, F., Conner, M., McMillan, B., & Ferguson, E. (2008). Effects of daily hassles and eating style on eating behavior. Health Psychology, 27(1 Suppl.), S20eS30.

Patel, K. A., & Schlundt, D. G. (2001). Impact of moods and social context on eating behavior. Appetite, 36(2), 111e118.

Pollard, T. M., Steptoe, A., Canaan, L., Davies, G. J., & Wardle, J. (1995). Effects of academic examination stress on eating behavior and blood lipid levels. Inter- national Journal of Behavioral Medicine, 2(4), 299e320.

Powers, S. I., Pietromonaco, P. R., Gunlicks, M., & Sayer, A. (2006). Dating couples' attachment styles and patterns of cortisol reactivity and recovery in response to a relationship conflict. Journal of Personality and Social Psychology, 90(4), 613e628.

Ross, S. E., Niebling, B. C., & Heckert, T. M. (1999). Sources of stress among college students. College Student Journal, 33(2), 312e317.

Shorey, H., & Snyder, C. (2006). The role of adult attachment styles in psychopa- thology and psychotherapy outcomes. Review of General Psychology, 10(1), 1e20.

Sijtsma, K. (2009). On the use, the misuse, and the very limited usefulness of Cronbach's alpha. Psychometrika, 74(1), 107e120.

Sinha, R., Catapano, D., & O'Malley, S. (1999). Stress-induced craving and stress

response in cocaine dependent individuals. Psychopharmacology, 142(4), 343e351.

Spangler, G., & Schieche, M. (1998). Emotional and adrenocortical responses of in- fants to the strange situation: the differential function of emotional expression. International Journal Behavioural Development, 22, 681e706.

Spielberger, C. D. (1989). State-trait anxiety inventory: Bibliography (2nd ed.). Palo Alto, CA: Consulting Psychologists Press.

Spielberger, C., Gorsuch, R., & Lushene, R. (1970). Manual for state-trait anxiety in- ventory. Palo Alto, CA: Consulting Psychologist Press.

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychol- ogists Press.

Stone, A. A., & Brownell, K. D. (1994). The stress eating paradox: multiple daily measurements in adult males and females. Psychology & Health, 9, 425e436.

Takai, N., Yamaguchi, M., Aragaki, T., Eto, K., Uchihashi, K., & Nishikawa, Y. (2004). Effect of psychological stress on the salivary cortisol and amylase levels in healthy young adults. Archives of Oral Biology, 49(12), 963e968.

Tasca, G. A., & Balfour, L. (2014). Attachment and eating disorders: a review of the current research. International Journal of Eating Disorders, 47(7), 710e717.

Tasca, G. A., Szadkowski, L., Illing, V., Trinneer, A., Grenon, R., Demidenko, N. … Bissada, H. (2009). Adult attachment, depression, and eating disorder symptoms: the mediating role of affect regulation strategies. Person- ality and Individual Differences, 47(6), 662e667.

Torres, S. J., & Nowson, C. A. (2007). Relationship between stress, eating behavior, and obesity. Nutrition, 23, 887e894.

Troisi, J., & Gabriel, S. (2014). Chicken soup really is good for the soul: “Comfort food” fulfills the need to belong. Psychological Science, 22, 747e753.

Ty, M., & Francis, A. J. (2013). Insecure attachment and disordered eating in women: the mediating processes of social comparison and emotion dysregulation. Eating Disorders, 21(2), 154e174.

Wallis, D. J., & Hetherington, M. M. (2004). Stress and eating: the effects of ego- threat and cognitive demand on food intake is restrained and emotional eaters. Appetite, 43(1), 39e46.

Wallis, D. J., & Hetherington, M. M. (2009). Emotions and eating. Self-reported and experimentally induced changes in food intake under stress. Appetite, 52, 355e362.

Wansink, B., Cheney, M. M., & Chan, N. (2003). Exploring comfort food preferences across age and gender. Physiology and Behaviour, 79, 739e747.

Wardle, J., Steptoe, A., Oliver, G., & Lipsey, Z. (2000). Stress, dietary restraint and food intake. Journal of Psychosomatic Research, 48, 195e202.

Ward, A., Ramsay, R., & Treasure, J. (2000). Attachment research in eating disorders. British Journal of Medical Psychology, 73(1), 35e51.

Weinstein, S. E., Shide, D. J., & Rolls, B. J. (1997). Changes in food intake in response to stress in men and women: psychological factors. Appetite, 28, 7e18.

Wilkinson, L., Rowe, A., Bishop, R. J., & Brunsrom, J. M. (2010). Attachment anxiety, disinhibited eating, and body mass index in adulthood. International Journal of Obesity, 34, 1442e1445.

Wilkinson, L., Rowe, A., & Heath, G. (2013). Eating me up inside: priming attach- ment security and anxiety, and their effects on snacking. Journal of Social and Personal Relationships, 30(6), 795e804.

Zellner, D. A., Loaiza, S., Gonzalex, Z., Pita, J., Morales, J., Pecora, D., et al. (2006). Food selection changes under stress. Physiology & Behaviour, 87, 789e793.