- Email: [email protected]
Effects of hunger and visuo-spatial interference on imagery-induced food cravings
Appetite 46 (2006) 36–40 www.elsevier.com/locate/appet
Research Report
Effects of hunger and visuo-spatial interference on imagery-induced food cravings Danielle Steel, Eva Kemps *, Marika Tiggemann School of Psychology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia Received 2 June 2005; received in revised form 19 October 2005; accepted 4 November 2005
Abstract The present study investigated the effects of hunger and visuo-spatial interference on imagery-induced food cravings. Forty-two women were randomly assigned to a hungry (no food for prior 4 h) or not hungry condition. Participants were asked to form and maintain images of desired foods while looking at a blank computer screen (control condition) or performing a task designed to load the visuo-spatial sketchpad of working memory (dynamic visual noise). They then rated the vividness of their images and their craving intensity. Although hungry participants reported stronger food cravings, dynamic visual noise made images less vivid and cravings less intense, irrespective of participant hunger status. Thus concurrent visuo-spatial processing may offer a useful technique for treating problematic food cravings that are predominantly psychological in origin, as well as those that are hunger-driven. q 2006 Elsevier Ltd. All rights reserved. Keywords: Food cravings; Hunger; Imagery; Working memory; Visuo-spatial sketchpad
Introduction Food cravings refer to intense desires or longings that are specific to the individual and that serve to motivate them to seek out and ingest a particular food (Cepeda-Benito and Gleaves, 2001). Such cravings are a common occurrence. For example, Weingarten and Elston (1991) reported that up to 97% of female and 68% of male college students experienced food cravings. Although food cravings are typically not pathological in nature, high levels of craving have been associated with dietary restraint, compulsive eating (Federoff, Polivy, & Herman, 2003; Green, 2001), obesity (Greeno, Wing, & Shiffman, 2000; Wurtman & Wurtman, 1995), and binge eating, especially in individuals with bulimia nervosa (Gendall, Joyce, Sullivan, & Bulik, 1998; Mitchell, Hatsukami, Eckert, & Pyle, 1985; Waters, Hill, & Waller, 2001). It is these potentially serious consequences that make investigation of the mechanisms underlying food craving crucial. While most researchers agree that cravings have at least some physiological basis, serving as a signal from the body that a nutritional deficiency and/or energy depletion needs to be * Corresponding author. E-mail address: [email protected] (E. Kemps).
0195-6663/$ - see front matter q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2005.11.001
redressed (Gibson & Desmond, 1999; Hill & Heaton-Brown, 1994; Weingarten & Elston, 1990), research has also established that hunger is not a necessary condition for the occurrence of food cravings. For example, Cornell, Rodin and Weingarten (1989) found that participants both reported cravings for and consumed pizza and ice cream upon presentation of these foods, irrespective of whether they were hungry or satiated. Similarly, Lambert, Nearl, Noyes, Parker, and Worrel (1991, 1992) have replicated this finding for chocolate. Thus physiological factors do not provide the entire origin of food cravings. Instead, the psychological literature cites an association between craving and negative mood states, such as depression (Dye, Warner, & Bancroft, 1995), anxiety (Jansen, 1998) and stress (Weingarten & Elston, 1991). Additionally, indulging a craving may initially produce a rush of positive affect, but is often followed by feelings of guilt and shame (Macdiarmid & Hetherington, 1995). More recent investigations from a cognitive perspective suggest that there is an imagery basis to food cravings. For example, Green, Rogers, and Elliman (2000) elicited food cravings in the laboratory via the use of imagery, and found that such cravings interfered with cognitive performance. Evidence from the surveys of May, Andrade, Panabokke, and Kavanagh (2004) and Tiggemann and Kemps (in press) corroborates that mental imagery is an important component of everyday cravings. In fact, Kavanagh, Andrade, and May (2005) have proposed a cognitive model of craving that
D. Steel et al. / Appetite 46 (2006) 36–40
assumes imagery to be at the core of the craving experience. In support, Harvey, Kemps, and Tiggemann (2005) document a positive relationship between the vividness of visual imagery and the intensity of the food craving experience. In recent years, imagery has been conceptualised within a working memory framework (Pearson, 2001). The most widely adopted model of working memory is that proposed by Baddeley and Hitch (1974) and Baddeley (2000), comprising a central executive and two limited-capacity subsystems, the visuo-spatial sketchpad and the phonological loop. The visuospatial sketchpad and phonological loop are responsible for the generation and maintenance of visual and auditory images, respectively. Because of their limited capacity, a reduction in processing efficiency occurs when additional load is placed upon the visuo-spatial sketchpad (phonological loop), such that information already in storage cannot be effectively retained when two tasks are competing for the same processing resources. Applying this reasoning, Kemps, Tiggemann, Woods, and Soekov (2004) demonstrated that loading the visuo-spatial sketchpad with a concurrent task while imagining popular foods (e.g. chocolate and cake) triggered by pictures reduced the vividness of participants’ food images, and correspondingly their level of craving. One such concurrent task was dynamic visual noise, which involved watching a flickering pattern of random black and white dots (Quinn & McConnell, 1996). Kemps, Tiggemann, and Hart (2005) subsequently showed that chocolate cravings were reduced more by concurrent dynamic visual noise than by a concurrent auditory task, demonstrating that interfering with the operation of the visuo-spatial sketchpad is a more effective way of reducing cravings than is interfering with the functioning of the phonological loop. An important issue that has not been considered in this context is hunger. In particular, it is not clear how much of the craving in the previous studies may have been due to hunger. Although there is some ambiguity in the general craving literature as to the precise role of hunger in food cravings (Cornell et al., 1989; Denton, 1984; Hill, Weaver, & Blundell, 1991; Kassel & Shiffman, 1992; Lambert et al., 1991, 1992), the distinction between cravings that are and are not derived from hunger may be a particularly important one, given that cravings arising for reasons other than hunger have the greater potential for problematic consequences. Accordingly, the current study sought to extend previous research by testing the efficacy of concurrent dynamic visual noise for reducing food cravings that were hunger driven as well as those that were not. The present study also sought to simulate a more naturalistic craving experience by asking participants to nominate their own craved foods, rather than using pictorial representations of specific yet commonly craved foods as in previous studies. It is possible that generally popular foods would not be craved by all individuals, and also that what is craved at any time may differ according to hunger status. Thus, stimuli were carefully tailored to the craving experience of the individual at the specific point in time. In addition, the current study included
37
measures of habitual food craving, restrained eating and general imaging ability as potential covariates. Method Participants Participants were 42 first year female psychology students from Flinders University, who volunteered to take part in the study in exchange for credit towards their degree. They ranged in age from 18 to 33 years. On registering interest, students were randomly assigned to either a hungry or not hungry condition and provided with the appropriate pre-testing preparatory instructions. Those in the hungry condition were asked to refrain from eating or drinking anything but water for at least 4 h prior to the testing session. Those in the not hungry condition were asked to eat immediately prior to the session. Initial hunger ratings (100 mm visual analogue scale) taken at the commencement of the testing session indicated two distinct groups, t(40)Z8.5, p!0.001. Those in the hungry condition reported much greater hunger (MZ60.1, SDZ23.4) than those in the not hungry condition (MZ10.1, SDZ13.4). Design The study utilised a 2!2 mixed factorial experimental design with the between-subjects factor of hunger status (2, hungry, not hungry) and the within-subjects factor of concurrent task (2, control, dynamic visual noise). Procedure Participants were tested individually, in a session lasting approximately 30 min. They first performed the imagery task and concurrent tasks and then completed self-report ratings of habitual food craving, restrained eating and imaging ability. Participants’ height and weight measurements were obtained and used to calculate body mass index (weight in kilogram/height in meter squared). Imagery task. Participants were asked to provide a list of the three foods that they would like to eat most ‘right now’. These foods then served as the stimuli for the imagery task. The most commonly reported foods were pasta (26%), chocolate (24%), and ice cream (11%). No difference in pattern was detected for the nominated foods by hunger status group. Participants provided baseline ratings of imagery vividness and craving intensity for each food using 100 mm visual analogue scales. For imagery vividness ratings, anchors ranged from ‘no image at all’ to ‘image perfectly clear—as vivid as normal vision’. For craving intensity ratings, the anchor points were ‘no desire or urge to eat this food’ to ‘extremely strong desire or urge to eat this food’. The baseline ratings of craving intensity provided a means of ranking the foods from most craved through to least craved, which were then used to counterbalance the order of foods and concurrent tasks for each trial. There was one trial for each of the three nominated foods per concurrent task condition.
38
D. Steel et al. / Appetite 46 (2006) 36–40
Concurrent task. For each concurrent task condition, participants were seated 45 cm in front of a 17-in. computer screen. On each trial participants were asked to form an image of one of their nominated foods for 5 s. They were asked to maintain that image for a further 8 s while performing a concurrent task. In the control condition, participants were required to focus on a blank computer screen. In the dynamic visual noise condition, participants focused on the computer screen that displayed a 17!17 cm matrix of random black and white squares, comprising 80 squares per row and column. The squares changed from black to white or white to black, at a rate of 640 changes per second, resulting in a ‘flickering’ effect. At the end of the retention interval, participants were asked to rate the vividness of their image and their craving intensity. Habitual food craving. The Food Cravings Questionnaire— Trait (Cepeda-Benito, Gleaves, Williams & Erath, 2000) was used to assess habitual food craving. The measure consists of 39 statements describing food cravings, such as ‘I feel like I have food on my mind all the time’. Participants rate the frequency of each description on a 6-point Likert scale ranging from 1 (‘never’) to 6 (‘always’). Total scores range from 39 to 234, with higher scores reflecting higher levels of habitual food craving. The authors report a test–retest reliability of 0.88 over three weeks. For the current sample, internal consistency was very high (Cronbach’s aZ0.95). Restrained eating. The Revised Restraint Scale (Herman & Polivy, 1980) was administered to assess chronic preoccupation with weight and dietary restraint. The measure consists of 10 items such as “How often are you dieting?” to which participants respond on a 4 or 5-point Likert scale. Total scores range from 0 to 35, with higher scores reflecting higher levels of dietary restraint in order to control weight. Internal consistency was moderate (aZ0.76) in the present sample. Imaging ability. The Vividness of Visual Imagery Questionnaire (Marks, 1973) was used to assess general imaging ability. The measure requires participants to imagine 16 familiar scenes, such as ‘A strong wind blows on the trees and on the lake causing waves’, and then to rate the vividness of the visual imagery that the scenes generated, on a 5-point scale ranging from 1 (perfectly clear and as vivid as normal vision) to 5 (no image at all, you only ‘know’ that you are thinking of the object). Total scores range from 16 to 80, with lower scores reflecting greater imaging ability. Internal consistency was high (aZ0.85).
Results Participant characteristics Table 1 presents means and standard deviations for participant characteristics. As can be seen, the two hunger status groups did not differ in terms of participant age or body mass index (BMI), nor on measures of habitual food craving, dietary restraint and general imaging ability. Thus there was no need to employ these as covariates in the subsequent analyses.
Table 1 Means for participant characteristics (standard deviations in parentheses)
Age BMI Habitual craving Restrained eating Imaging ability
Hungry
Not hungry
22.0 (4.3) 23.4 (5.4) 122.8 (32.3) 13.7 (4.8) 40.2 (9.5)
21.5 (4.2) 22.5 (4.4) 131.5 (34.3) 13.5 (5.5) 36.7 (9.6)
Imagery task As there was no significant difference in imagery vividness or craving intensity ratings between the 3 trials of each concurrent task condition (Fs!1), ratings were averaged across trials. These ratings were analysed by 2 (hunger status)!2 (concurrent task) mixed design ANOVAs. Imagery vividness. There was no significant main effect of hunger status, F(1, 40)Z2.0, pO0.15. There was, however, a significant main effect of concurrent task, F(1, 40)Z51.7, p!0.001, whereby imagery vividness ratings were lower for the dynamic visual noise condition (MZ44.0) than for the control condition (MZ72.8). There was also a marginally significant interaction between these two factors, F(1, 40)Z3.7, pZ0.06. While imagery vividness ratings in the dynamic visual noise condition for both groups decreased relative to the control condition, there was a considerably greater decrease for those in the hungry condition (50%) than for those in the not hungry condition (29%). The upper part of Table 2 clearly shows that there was no difference between hunger status groups in the control condition, t(40)Z0.1, pO0.9, but that the not hungry group had significantly lower vividness ratings in the dynamic visual noise condition, t(40)Z2.1, p!0.05. Craving intensity. There was a significant main effect of hunger status, F(1, 40)Z21.0, p!0.001, with those in the hungry group reporting more intense cravings (MZ58.6) than those in the not hungry group (MZ32.7). There was also a significant main effect of concurrent task, F(1, 40)Z42.6, p!0.001, whereby craving intensity ratings were lower for the dynamic visual noise condition (MZ36.6) than for the control condition (MZ54.4). There was no significant interaction effect between concurrent task and hunger status, F!1. The lower part of Table 2 presents the craving intensity ratings by hunger status group and concurrent task condition. Relationship between imagery vividness and craving intensity. Pearson product–moment correlations were conducted using the averaged ratings of imagery vividness and Table 2 Mean imagery vividness and craving intensity ratings as a function of hunger status and concurrent task (standard deviations in parentheses) Ratings
Concurrent task
Hunger status Hungry
Not hungry
Imagery vividness
Control Dynamic visual noise Control Dynamic visual noise
73.2 (21.0) 52.1 (25.5) 66.7 (18.1) 50.6 (22.5)
72.4 (21.9) 35.9 (25.7) 42.1 (24.1) 22.7 (16.6)
Craving intensity
D. Steel et al. / Appetite 46 (2006) 36–40
craving intensity. Results revealed a significant correlation, r(42)Z0.60, p!0.001, with higher ratings of imagery vividness associated with higher ratings of craving intensity. When the same analyses were conducted for each cell of the design, significant correlations were found between vividness and craving ratings for the hungry group [control: r(21)Z0.59, p!0.01; dynamic visual noise: r(21)Z0.82, p!0.001] and the not hungry group [control: r(21)Z0.56, p!0.01; dynamic visual noise: r(21)Z0.57, p!0.01]. Discussion The current study aimed to extend previous research using a working memory paradigm to reduce food cravings. It examined whether concurrent visuo-spatial processing would differentially affect cravings that arise from hunger and those that are not hunger driven. This was tested using participantgenerated rather than experimenter—provided foods as stimuli for eliciting cravings. Irrespective of hunger status, participants most frequently listed pasta and chocolate as the foods they most wanted to eat at the start of the testing session. This is consistent with the literature that suggests that women most commonly crave carbohydrates (Buffenstein, Poppitt, McDevitt, & Prentice, 1995; Dye et al., 1995). Also in support of some previous research demonstrating an association between hunger and food cravings (Denton, 1984; Hill et al., 1991; Kassel & Shiffman, 1992), hungry participants reported more intense cravings than did non-hungry participants. This finding is in line with the classic starvation study of Keys, Brozek, Henschel, Mickelsen, and Taylor (1950) who observed that food cravings did not abate the longer that participants went without food. Of central importance, however, we found that performing a concurrent task that loads the visuo-spatial sketchpad served to reduce not only the vividness with which the food image was experienced, but also the associated craving for that food. According to the working memory paradigm, this reduction comes about because concurrent visual–spatial processing competes for limited capacity in the visuo-spatial working memory resource. Thus we have replicated the earlier findings of Kemps and colleagues (2004, 2005) and extended them to cravings for foods nominated by participants. Importantly, the reduction in imagery vividness and craving intensity occurred irrespective of whether cravings arose as a result of hunger or for other reasons. Thus the technique seems quite robust. Further, the underlying mechanism appears to operate in the same way for hungry and non-hungry participants. Although the magnitude of reduction in vividness of imagery was greater in the not hungry group, the vividness of the food image was correlated with craving intensity for both hunger status groups. The finding that dynamic visual noise reduced both hungerdriven and non-hunger-driven cravings has several important practical implications. In essence, cravings triggered by hunger should not be problematic. They simply signal a physiological need to eat. However, even cravings that are hunger-driven may pose problems for some people, for example, overweight
39
individuals dieting to lose weight. This may be particularly pertinent for medically prescribed diets or food restrictions, as in the case of individuals suffering from obesity, cardiovascular disease or Type II diabetes. Because cravings have been linked to early dropout from weight-loss programs (Sitton, 1991), simple visuo-spatial techniques like dynamic visual noise may help to alleviate hunger-driven cravings and make it easier for individuals to adhere to their prescribed diet. Perhaps more useful, dynamic visual noise was shown to work equally well for cravings triggered by factors other than hunger. In general, these are likely to have more negative consequences. Overweight and obese individuals are often motivated to eat for reasons other than the signals of hunger and satiety (Schacter & Rodin, 1974). Cravings arising from negative mood states and environmental cues have also been associated with binge eating in individuals with bulimia nervosa (Gendall et al., 1998; Mitchell et al., 1985; Waters et al., 2001) and obesity (Greeno et al., 2000; Wurtman & Wurtman, 1986). Hence, concurrent visuo-spatial processing may offer a simple but useful technique for reducing the intensity of psychologically-induced cravings in obese and bulimic individuals, and thereby help curb unwanted eating or overeating. In sum, the current study demonstrates the efficacy of visuospatial techniques for the reduction of food cravings, irrespective of hunger level. Future research needs to extend these findings to samples for whom food craving represents a serious problem. In particular, we need to determine whether these techniques can benefit individuals who are overweight, obese or suffering from bulimia or binge eating disorder. References Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4, 417–423. Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. A. Bower (Ed.), Recent advances in learning and motivation (pp. 47–89). New York: Academic Press. Buffenstein, R., Poppitt, S. D., McDevitt, R. M., & Prentice, A. M. (1995). Food intake and the menstrual cycle: A retrospective analysis, with implications for appetite research. Physiology and Behavior, 58, 1067– 1077. Cepeda-Benito, A., & Gleaves, D. H. (2001). A critique of food cravings research: Theory, measurement, and food intake. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 3–27). Surrey, UK: Leatherhead Publishing. Cepeda-Benito, A., Gleaves, D. H., Williams, T. L., & Erath, S. A. (2000). The development and validation of the state and trait food-cravings questionnaires. Behavior Therapy, 31, 151–173. Cornell, C. E., Rodin, J., & Weingarten, H. (1989). Stimulus-induced eating when satiated. Physiology and Behavior, 45, 695–704. Denton, D. (1984). The hunger for salt. New York: Springer. Dye, L., Warner, P., & Bancroft, J. (1995). Food craving and the menstrual cycle and its relationship to stress, happiness of relationship and depression: A preliminary enquiry. Journal of Affective Disorders, 34, 157–164. Fedoroff, I., Polivy, J., & Herman, C. P. (2003). The specificity of restrained versus unrestrained eaters’ responses to food cues: General desire to eat, or craving the cued food? Appetite, 41, 7–13. Gendall, K. A., Joyce, P. R., Sullivan, P. F., & Bulik, C. M. (1998). Food cravers: Characteristics of those who binge. The International Journal of Eating Disorders, 23, 353–360.
40
D. Steel et al. / Appetite 46 (2006) 36–40
Gibson, E. L., & Desmond, E. (1999). Chocolate craving and hunger state: Implications for the acquisition and expression of appetite and food choice. Appetite, 32, 219–240. Green, M. (2001). Dietary restraint and craving. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 521–548). Surrey, UK: Leatherhead Publishing. Green, M. W., Rogers, P. J., & Elliman, N. A. (2000). Dietary restraint and addictive behaviours: The generalisability of Tiffany’s cue reactivity model. The International Journal of Eating Disorders, 27, 419–427. Greeno, C. G., Wing, R. R., & Schiffman, S. (2000). Binge antecedents in obese women with and without binge eating disorder. Journal of Consulting and Clinical Psychology, 68, 95–102. Harvey, K., Kemps, E., & Tiggemann, M. (2005). The nature of imagery processes underlying food cravings. British Journal of Health Psychology, 10, 49–56. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity (pp. 208–225). Philadelphia, PA: W.B. Saunders. Hill, A. J., & Heaton-Brown, L. (1994). The experience of food craving: A prospective investigation in healthy women. Journal of Psychosomatic Research, 38, 801–814. Hill, A. J., Weaver, C. F. L., & Blundell, J. E. (1991). Food craving, dietary restraint and mood. Appetite, 17, 187–197. Jansen, A. (1998). A learning model of binge eating: Cue reactivity and cue exposure. Behaviour Research and Therapy, 36, 257–272. Kassel, J. D., & Shiffman, S. (1992). What can hunger teach us about drug craving? A comparative analysis of the two constructs. Advances in Behaviour Therapy and Research, 14, 141–167. Kavanagh, D. J., Andrade, J., & May, J. (2005). Imaginary relish and exquisite torture: The elaborated intrusion theory of desire. Psychological Review, 112, 446–467. Kemps, E., Tiggemann, M., & Hart, G. (2005). Chocolate cravings are susceptible to visuo-spatial interference. Eating Behaviors, 6, 101–107. Kemps, E., Tiggemann, M., Woods, D., & Soekov, B. (2004). Reduction of food cravings through concurrent visuo-spatial processing. The International Journal of Eating Disorders, 36, 31–40. Keys, A., Brozek, J., Henschel, A., Mickelsen, O., & Taylor, H. L. (1950). The biology of human starvation. Minneapolis, MN: University of Minnesota Press.
Lambert, K. G., Neal, T., Noyes, J., Parker, K., & Worrel, P. (1991–1992). Food-related stimuli increase desire to eat in hungry and satiated human subjects. Current Psychology: Research and Reviews, 10, 297–303. Macdiarmid, J. I., & Hetherington, M. M. (1995). Mood modulation by food: An exploration of affect on cravings in ‘chocolate addicts’. The British Journal of Clinical Psychology, 34, 129–138. Marks, D. F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology, 64, 17–24. May, J., Andrade, J., Panabokke, N., & Kavanagh, D. (2004). Images of desire: Cognitive models of craving. Memory, 12, 447–461. Mitchell, J. E., Hatsukami, D., Eckert, E. D., & Pyle, R. L. (1985). Characteristics of 275 patients with bulimia. The American Journal of Psychiatry, 142, 482–485. Pearson, D. G. (2001). Imagery and the visuo-spatial sketchpad. In J. Andrade (Ed.), Working memory in perspective (pp. 33–59). Hove, UK: Taylor & Francis. Quinn, J. G., & McConnell, J. (1996). Irrelevant pictures in visual working memory. The Quarterly Journal of Experimental Psychology, 49A, 200–215. Schacter, S., & Rodin, J. (1974). Obese humans and rats. Washington, DC: Erlbaum. Sitton, S. C. (1991). Role of craving for carbohydrates upon completion of a protein-sparing fast. Psychological Reports, 69, 683–686. Tiggemann, M., & Kemps, E. (2005). The phenomenology of food cravings: The role of mental imagery. Appetite, 45, 305–313. Waters, A., Hill, A., & Waller, G. (2001). Bulimics responses to food cravings: Is binge-eating a product of hunger or emotional state? Behaviour Research and Therapy, 39, 877–886. Weingarten, H. P., & Elston, D. (1990). The phenomenology of food cravings. Appetite, 15, 231–246. Weingarten, H. P., & Elston, D. (1991). Food cravings in a college population. Appetite, 17, 167–175. Wurtman, R. J., & Wurtman, J. J. (1986). Carbohydrate craving, obesity and brain serotonin. Appetite, 7, 99–103. Wurtman, R. J., & Wurtman, J. J. (1995). Brain serotonin, carbohydrate craving, obesity and depression. Obesity Research, 3, 477–480.
Research Report
Effects of hunger and visuo-spatial interference on imagery-induced food cravings Danielle Steel, Eva Kemps *, Marika Tiggemann School of Psychology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia Received 2 June 2005; received in revised form 19 October 2005; accepted 4 November 2005
Abstract The present study investigated the effects of hunger and visuo-spatial interference on imagery-induced food cravings. Forty-two women were randomly assigned to a hungry (no food for prior 4 h) or not hungry condition. Participants were asked to form and maintain images of desired foods while looking at a blank computer screen (control condition) or performing a task designed to load the visuo-spatial sketchpad of working memory (dynamic visual noise). They then rated the vividness of their images and their craving intensity. Although hungry participants reported stronger food cravings, dynamic visual noise made images less vivid and cravings less intense, irrespective of participant hunger status. Thus concurrent visuo-spatial processing may offer a useful technique for treating problematic food cravings that are predominantly psychological in origin, as well as those that are hunger-driven. q 2006 Elsevier Ltd. All rights reserved. Keywords: Food cravings; Hunger; Imagery; Working memory; Visuo-spatial sketchpad
Introduction Food cravings refer to intense desires or longings that are specific to the individual and that serve to motivate them to seek out and ingest a particular food (Cepeda-Benito and Gleaves, 2001). Such cravings are a common occurrence. For example, Weingarten and Elston (1991) reported that up to 97% of female and 68% of male college students experienced food cravings. Although food cravings are typically not pathological in nature, high levels of craving have been associated with dietary restraint, compulsive eating (Federoff, Polivy, & Herman, 2003; Green, 2001), obesity (Greeno, Wing, & Shiffman, 2000; Wurtman & Wurtman, 1995), and binge eating, especially in individuals with bulimia nervosa (Gendall, Joyce, Sullivan, & Bulik, 1998; Mitchell, Hatsukami, Eckert, & Pyle, 1985; Waters, Hill, & Waller, 2001). It is these potentially serious consequences that make investigation of the mechanisms underlying food craving crucial. While most researchers agree that cravings have at least some physiological basis, serving as a signal from the body that a nutritional deficiency and/or energy depletion needs to be * Corresponding author. E-mail address: [email protected] (E. Kemps).
0195-6663/$ - see front matter q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2005.11.001
redressed (Gibson & Desmond, 1999; Hill & Heaton-Brown, 1994; Weingarten & Elston, 1990), research has also established that hunger is not a necessary condition for the occurrence of food cravings. For example, Cornell, Rodin and Weingarten (1989) found that participants both reported cravings for and consumed pizza and ice cream upon presentation of these foods, irrespective of whether they were hungry or satiated. Similarly, Lambert, Nearl, Noyes, Parker, and Worrel (1991, 1992) have replicated this finding for chocolate. Thus physiological factors do not provide the entire origin of food cravings. Instead, the psychological literature cites an association between craving and negative mood states, such as depression (Dye, Warner, & Bancroft, 1995), anxiety (Jansen, 1998) and stress (Weingarten & Elston, 1991). Additionally, indulging a craving may initially produce a rush of positive affect, but is often followed by feelings of guilt and shame (Macdiarmid & Hetherington, 1995). More recent investigations from a cognitive perspective suggest that there is an imagery basis to food cravings. For example, Green, Rogers, and Elliman (2000) elicited food cravings in the laboratory via the use of imagery, and found that such cravings interfered with cognitive performance. Evidence from the surveys of May, Andrade, Panabokke, and Kavanagh (2004) and Tiggemann and Kemps (in press) corroborates that mental imagery is an important component of everyday cravings. In fact, Kavanagh, Andrade, and May (2005) have proposed a cognitive model of craving that
D. Steel et al. / Appetite 46 (2006) 36–40
assumes imagery to be at the core of the craving experience. In support, Harvey, Kemps, and Tiggemann (2005) document a positive relationship between the vividness of visual imagery and the intensity of the food craving experience. In recent years, imagery has been conceptualised within a working memory framework (Pearson, 2001). The most widely adopted model of working memory is that proposed by Baddeley and Hitch (1974) and Baddeley (2000), comprising a central executive and two limited-capacity subsystems, the visuo-spatial sketchpad and the phonological loop. The visuospatial sketchpad and phonological loop are responsible for the generation and maintenance of visual and auditory images, respectively. Because of their limited capacity, a reduction in processing efficiency occurs when additional load is placed upon the visuo-spatial sketchpad (phonological loop), such that information already in storage cannot be effectively retained when two tasks are competing for the same processing resources. Applying this reasoning, Kemps, Tiggemann, Woods, and Soekov (2004) demonstrated that loading the visuo-spatial sketchpad with a concurrent task while imagining popular foods (e.g. chocolate and cake) triggered by pictures reduced the vividness of participants’ food images, and correspondingly their level of craving. One such concurrent task was dynamic visual noise, which involved watching a flickering pattern of random black and white dots (Quinn & McConnell, 1996). Kemps, Tiggemann, and Hart (2005) subsequently showed that chocolate cravings were reduced more by concurrent dynamic visual noise than by a concurrent auditory task, demonstrating that interfering with the operation of the visuo-spatial sketchpad is a more effective way of reducing cravings than is interfering with the functioning of the phonological loop. An important issue that has not been considered in this context is hunger. In particular, it is not clear how much of the craving in the previous studies may have been due to hunger. Although there is some ambiguity in the general craving literature as to the precise role of hunger in food cravings (Cornell et al., 1989; Denton, 1984; Hill, Weaver, & Blundell, 1991; Kassel & Shiffman, 1992; Lambert et al., 1991, 1992), the distinction between cravings that are and are not derived from hunger may be a particularly important one, given that cravings arising for reasons other than hunger have the greater potential for problematic consequences. Accordingly, the current study sought to extend previous research by testing the efficacy of concurrent dynamic visual noise for reducing food cravings that were hunger driven as well as those that were not. The present study also sought to simulate a more naturalistic craving experience by asking participants to nominate their own craved foods, rather than using pictorial representations of specific yet commonly craved foods as in previous studies. It is possible that generally popular foods would not be craved by all individuals, and also that what is craved at any time may differ according to hunger status. Thus, stimuli were carefully tailored to the craving experience of the individual at the specific point in time. In addition, the current study included
37
measures of habitual food craving, restrained eating and general imaging ability as potential covariates. Method Participants Participants were 42 first year female psychology students from Flinders University, who volunteered to take part in the study in exchange for credit towards their degree. They ranged in age from 18 to 33 years. On registering interest, students were randomly assigned to either a hungry or not hungry condition and provided with the appropriate pre-testing preparatory instructions. Those in the hungry condition were asked to refrain from eating or drinking anything but water for at least 4 h prior to the testing session. Those in the not hungry condition were asked to eat immediately prior to the session. Initial hunger ratings (100 mm visual analogue scale) taken at the commencement of the testing session indicated two distinct groups, t(40)Z8.5, p!0.001. Those in the hungry condition reported much greater hunger (MZ60.1, SDZ23.4) than those in the not hungry condition (MZ10.1, SDZ13.4). Design The study utilised a 2!2 mixed factorial experimental design with the between-subjects factor of hunger status (2, hungry, not hungry) and the within-subjects factor of concurrent task (2, control, dynamic visual noise). Procedure Participants were tested individually, in a session lasting approximately 30 min. They first performed the imagery task and concurrent tasks and then completed self-report ratings of habitual food craving, restrained eating and imaging ability. Participants’ height and weight measurements were obtained and used to calculate body mass index (weight in kilogram/height in meter squared). Imagery task. Participants were asked to provide a list of the three foods that they would like to eat most ‘right now’. These foods then served as the stimuli for the imagery task. The most commonly reported foods were pasta (26%), chocolate (24%), and ice cream (11%). No difference in pattern was detected for the nominated foods by hunger status group. Participants provided baseline ratings of imagery vividness and craving intensity for each food using 100 mm visual analogue scales. For imagery vividness ratings, anchors ranged from ‘no image at all’ to ‘image perfectly clear—as vivid as normal vision’. For craving intensity ratings, the anchor points were ‘no desire or urge to eat this food’ to ‘extremely strong desire or urge to eat this food’. The baseline ratings of craving intensity provided a means of ranking the foods from most craved through to least craved, which were then used to counterbalance the order of foods and concurrent tasks for each trial. There was one trial for each of the three nominated foods per concurrent task condition.
38
D. Steel et al. / Appetite 46 (2006) 36–40
Concurrent task. For each concurrent task condition, participants were seated 45 cm in front of a 17-in. computer screen. On each trial participants were asked to form an image of one of their nominated foods for 5 s. They were asked to maintain that image for a further 8 s while performing a concurrent task. In the control condition, participants were required to focus on a blank computer screen. In the dynamic visual noise condition, participants focused on the computer screen that displayed a 17!17 cm matrix of random black and white squares, comprising 80 squares per row and column. The squares changed from black to white or white to black, at a rate of 640 changes per second, resulting in a ‘flickering’ effect. At the end of the retention interval, participants were asked to rate the vividness of their image and their craving intensity. Habitual food craving. The Food Cravings Questionnaire— Trait (Cepeda-Benito, Gleaves, Williams & Erath, 2000) was used to assess habitual food craving. The measure consists of 39 statements describing food cravings, such as ‘I feel like I have food on my mind all the time’. Participants rate the frequency of each description on a 6-point Likert scale ranging from 1 (‘never’) to 6 (‘always’). Total scores range from 39 to 234, with higher scores reflecting higher levels of habitual food craving. The authors report a test–retest reliability of 0.88 over three weeks. For the current sample, internal consistency was very high (Cronbach’s aZ0.95). Restrained eating. The Revised Restraint Scale (Herman & Polivy, 1980) was administered to assess chronic preoccupation with weight and dietary restraint. The measure consists of 10 items such as “How often are you dieting?” to which participants respond on a 4 or 5-point Likert scale. Total scores range from 0 to 35, with higher scores reflecting higher levels of dietary restraint in order to control weight. Internal consistency was moderate (aZ0.76) in the present sample. Imaging ability. The Vividness of Visual Imagery Questionnaire (Marks, 1973) was used to assess general imaging ability. The measure requires participants to imagine 16 familiar scenes, such as ‘A strong wind blows on the trees and on the lake causing waves’, and then to rate the vividness of the visual imagery that the scenes generated, on a 5-point scale ranging from 1 (perfectly clear and as vivid as normal vision) to 5 (no image at all, you only ‘know’ that you are thinking of the object). Total scores range from 16 to 80, with lower scores reflecting greater imaging ability. Internal consistency was high (aZ0.85).
Results Participant characteristics Table 1 presents means and standard deviations for participant characteristics. As can be seen, the two hunger status groups did not differ in terms of participant age or body mass index (BMI), nor on measures of habitual food craving, dietary restraint and general imaging ability. Thus there was no need to employ these as covariates in the subsequent analyses.
Table 1 Means for participant characteristics (standard deviations in parentheses)
Age BMI Habitual craving Restrained eating Imaging ability
Hungry
Not hungry
22.0 (4.3) 23.4 (5.4) 122.8 (32.3) 13.7 (4.8) 40.2 (9.5)
21.5 (4.2) 22.5 (4.4) 131.5 (34.3) 13.5 (5.5) 36.7 (9.6)
Imagery task As there was no significant difference in imagery vividness or craving intensity ratings between the 3 trials of each concurrent task condition (Fs!1), ratings were averaged across trials. These ratings were analysed by 2 (hunger status)!2 (concurrent task) mixed design ANOVAs. Imagery vividness. There was no significant main effect of hunger status, F(1, 40)Z2.0, pO0.15. There was, however, a significant main effect of concurrent task, F(1, 40)Z51.7, p!0.001, whereby imagery vividness ratings were lower for the dynamic visual noise condition (MZ44.0) than for the control condition (MZ72.8). There was also a marginally significant interaction between these two factors, F(1, 40)Z3.7, pZ0.06. While imagery vividness ratings in the dynamic visual noise condition for both groups decreased relative to the control condition, there was a considerably greater decrease for those in the hungry condition (50%) than for those in the not hungry condition (29%). The upper part of Table 2 clearly shows that there was no difference between hunger status groups in the control condition, t(40)Z0.1, pO0.9, but that the not hungry group had significantly lower vividness ratings in the dynamic visual noise condition, t(40)Z2.1, p!0.05. Craving intensity. There was a significant main effect of hunger status, F(1, 40)Z21.0, p!0.001, with those in the hungry group reporting more intense cravings (MZ58.6) than those in the not hungry group (MZ32.7). There was also a significant main effect of concurrent task, F(1, 40)Z42.6, p!0.001, whereby craving intensity ratings were lower for the dynamic visual noise condition (MZ36.6) than for the control condition (MZ54.4). There was no significant interaction effect between concurrent task and hunger status, F!1. The lower part of Table 2 presents the craving intensity ratings by hunger status group and concurrent task condition. Relationship between imagery vividness and craving intensity. Pearson product–moment correlations were conducted using the averaged ratings of imagery vividness and Table 2 Mean imagery vividness and craving intensity ratings as a function of hunger status and concurrent task (standard deviations in parentheses) Ratings
Concurrent task
Hunger status Hungry
Not hungry
Imagery vividness
Control Dynamic visual noise Control Dynamic visual noise
73.2 (21.0) 52.1 (25.5) 66.7 (18.1) 50.6 (22.5)
72.4 (21.9) 35.9 (25.7) 42.1 (24.1) 22.7 (16.6)
Craving intensity
D. Steel et al. / Appetite 46 (2006) 36–40
craving intensity. Results revealed a significant correlation, r(42)Z0.60, p!0.001, with higher ratings of imagery vividness associated with higher ratings of craving intensity. When the same analyses were conducted for each cell of the design, significant correlations were found between vividness and craving ratings for the hungry group [control: r(21)Z0.59, p!0.01; dynamic visual noise: r(21)Z0.82, p!0.001] and the not hungry group [control: r(21)Z0.56, p!0.01; dynamic visual noise: r(21)Z0.57, p!0.01]. Discussion The current study aimed to extend previous research using a working memory paradigm to reduce food cravings. It examined whether concurrent visuo-spatial processing would differentially affect cravings that arise from hunger and those that are not hunger driven. This was tested using participantgenerated rather than experimenter—provided foods as stimuli for eliciting cravings. Irrespective of hunger status, participants most frequently listed pasta and chocolate as the foods they most wanted to eat at the start of the testing session. This is consistent with the literature that suggests that women most commonly crave carbohydrates (Buffenstein, Poppitt, McDevitt, & Prentice, 1995; Dye et al., 1995). Also in support of some previous research demonstrating an association between hunger and food cravings (Denton, 1984; Hill et al., 1991; Kassel & Shiffman, 1992), hungry participants reported more intense cravings than did non-hungry participants. This finding is in line with the classic starvation study of Keys, Brozek, Henschel, Mickelsen, and Taylor (1950) who observed that food cravings did not abate the longer that participants went without food. Of central importance, however, we found that performing a concurrent task that loads the visuo-spatial sketchpad served to reduce not only the vividness with which the food image was experienced, but also the associated craving for that food. According to the working memory paradigm, this reduction comes about because concurrent visual–spatial processing competes for limited capacity in the visuo-spatial working memory resource. Thus we have replicated the earlier findings of Kemps and colleagues (2004, 2005) and extended them to cravings for foods nominated by participants. Importantly, the reduction in imagery vividness and craving intensity occurred irrespective of whether cravings arose as a result of hunger or for other reasons. Thus the technique seems quite robust. Further, the underlying mechanism appears to operate in the same way for hungry and non-hungry participants. Although the magnitude of reduction in vividness of imagery was greater in the not hungry group, the vividness of the food image was correlated with craving intensity for both hunger status groups. The finding that dynamic visual noise reduced both hungerdriven and non-hunger-driven cravings has several important practical implications. In essence, cravings triggered by hunger should not be problematic. They simply signal a physiological need to eat. However, even cravings that are hunger-driven may pose problems for some people, for example, overweight
39
individuals dieting to lose weight. This may be particularly pertinent for medically prescribed diets or food restrictions, as in the case of individuals suffering from obesity, cardiovascular disease or Type II diabetes. Because cravings have been linked to early dropout from weight-loss programs (Sitton, 1991), simple visuo-spatial techniques like dynamic visual noise may help to alleviate hunger-driven cravings and make it easier for individuals to adhere to their prescribed diet. Perhaps more useful, dynamic visual noise was shown to work equally well for cravings triggered by factors other than hunger. In general, these are likely to have more negative consequences. Overweight and obese individuals are often motivated to eat for reasons other than the signals of hunger and satiety (Schacter & Rodin, 1974). Cravings arising from negative mood states and environmental cues have also been associated with binge eating in individuals with bulimia nervosa (Gendall et al., 1998; Mitchell et al., 1985; Waters et al., 2001) and obesity (Greeno et al., 2000; Wurtman & Wurtman, 1986). Hence, concurrent visuo-spatial processing may offer a simple but useful technique for reducing the intensity of psychologically-induced cravings in obese and bulimic individuals, and thereby help curb unwanted eating or overeating. In sum, the current study demonstrates the efficacy of visuospatial techniques for the reduction of food cravings, irrespective of hunger level. Future research needs to extend these findings to samples for whom food craving represents a serious problem. In particular, we need to determine whether these techniques can benefit individuals who are overweight, obese or suffering from bulimia or binge eating disorder. References Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4, 417–423. Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. A. Bower (Ed.), Recent advances in learning and motivation (pp. 47–89). New York: Academic Press. Buffenstein, R., Poppitt, S. D., McDevitt, R. M., & Prentice, A. M. (1995). Food intake and the menstrual cycle: A retrospective analysis, with implications for appetite research. Physiology and Behavior, 58, 1067– 1077. Cepeda-Benito, A., & Gleaves, D. H. (2001). A critique of food cravings research: Theory, measurement, and food intake. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 3–27). Surrey, UK: Leatherhead Publishing. Cepeda-Benito, A., Gleaves, D. H., Williams, T. L., & Erath, S. A. (2000). The development and validation of the state and trait food-cravings questionnaires. Behavior Therapy, 31, 151–173. Cornell, C. E., Rodin, J., & Weingarten, H. (1989). Stimulus-induced eating when satiated. Physiology and Behavior, 45, 695–704. Denton, D. (1984). The hunger for salt. New York: Springer. Dye, L., Warner, P., & Bancroft, J. (1995). Food craving and the menstrual cycle and its relationship to stress, happiness of relationship and depression: A preliminary enquiry. Journal of Affective Disorders, 34, 157–164. Fedoroff, I., Polivy, J., & Herman, C. P. (2003). The specificity of restrained versus unrestrained eaters’ responses to food cues: General desire to eat, or craving the cued food? Appetite, 41, 7–13. Gendall, K. A., Joyce, P. R., Sullivan, P. F., & Bulik, C. M. (1998). Food cravers: Characteristics of those who binge. The International Journal of Eating Disorders, 23, 353–360.
40
D. Steel et al. / Appetite 46 (2006) 36–40
Gibson, E. L., & Desmond, E. (1999). Chocolate craving and hunger state: Implications for the acquisition and expression of appetite and food choice. Appetite, 32, 219–240. Green, M. (2001). Dietary restraint and craving. In M. M. Hetherington (Ed.), Food cravings and addiction (pp. 521–548). Surrey, UK: Leatherhead Publishing. Green, M. W., Rogers, P. J., & Elliman, N. A. (2000). Dietary restraint and addictive behaviours: The generalisability of Tiffany’s cue reactivity model. The International Journal of Eating Disorders, 27, 419–427. Greeno, C. G., Wing, R. R., & Schiffman, S. (2000). Binge antecedents in obese women with and without binge eating disorder. Journal of Consulting and Clinical Psychology, 68, 95–102. Harvey, K., Kemps, E., & Tiggemann, M. (2005). The nature of imagery processes underlying food cravings. British Journal of Health Psychology, 10, 49–56. Herman, C. P., & Polivy, J. (1980). Restrained eating. In A. J. Stunkard (Ed.), Obesity (pp. 208–225). Philadelphia, PA: W.B. Saunders. Hill, A. J., & Heaton-Brown, L. (1994). The experience of food craving: A prospective investigation in healthy women. Journal of Psychosomatic Research, 38, 801–814. Hill, A. J., Weaver, C. F. L., & Blundell, J. E. (1991). Food craving, dietary restraint and mood. Appetite, 17, 187–197. Jansen, A. (1998). A learning model of binge eating: Cue reactivity and cue exposure. Behaviour Research and Therapy, 36, 257–272. Kassel, J. D., & Shiffman, S. (1992). What can hunger teach us about drug craving? A comparative analysis of the two constructs. Advances in Behaviour Therapy and Research, 14, 141–167. Kavanagh, D. J., Andrade, J., & May, J. (2005). Imaginary relish and exquisite torture: The elaborated intrusion theory of desire. Psychological Review, 112, 446–467. Kemps, E., Tiggemann, M., & Hart, G. (2005). Chocolate cravings are susceptible to visuo-spatial interference. Eating Behaviors, 6, 101–107. Kemps, E., Tiggemann, M., Woods, D., & Soekov, B. (2004). Reduction of food cravings through concurrent visuo-spatial processing. The International Journal of Eating Disorders, 36, 31–40. Keys, A., Brozek, J., Henschel, A., Mickelsen, O., & Taylor, H. L. (1950). The biology of human starvation. Minneapolis, MN: University of Minnesota Press.
Lambert, K. G., Neal, T., Noyes, J., Parker, K., & Worrel, P. (1991–1992). Food-related stimuli increase desire to eat in hungry and satiated human subjects. Current Psychology: Research and Reviews, 10, 297–303. Macdiarmid, J. I., & Hetherington, M. M. (1995). Mood modulation by food: An exploration of affect on cravings in ‘chocolate addicts’. The British Journal of Clinical Psychology, 34, 129–138. Marks, D. F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology, 64, 17–24. May, J., Andrade, J., Panabokke, N., & Kavanagh, D. (2004). Images of desire: Cognitive models of craving. Memory, 12, 447–461. Mitchell, J. E., Hatsukami, D., Eckert, E. D., & Pyle, R. L. (1985). Characteristics of 275 patients with bulimia. The American Journal of Psychiatry, 142, 482–485. Pearson, D. G. (2001). Imagery and the visuo-spatial sketchpad. In J. Andrade (Ed.), Working memory in perspective (pp. 33–59). Hove, UK: Taylor & Francis. Quinn, J. G., & McConnell, J. (1996). Irrelevant pictures in visual working memory. The Quarterly Journal of Experimental Psychology, 49A, 200–215. Schacter, S., & Rodin, J. (1974). Obese humans and rats. Washington, DC: Erlbaum. Sitton, S. C. (1991). Role of craving for carbohydrates upon completion of a protein-sparing fast. Psychological Reports, 69, 683–686. Tiggemann, M., & Kemps, E. (2005). The phenomenology of food cravings: The role of mental imagery. Appetite, 45, 305–313. Waters, A., Hill, A., & Waller, G. (2001). Bulimics responses to food cravings: Is binge-eating a product of hunger or emotional state? Behaviour Research and Therapy, 39, 877–886. Weingarten, H. P., & Elston, D. (1990). The phenomenology of food cravings. Appetite, 15, 231–246. Weingarten, H. P., & Elston, D. (1991). Food cravings in a college population. Appetite, 17, 167–175. Wurtman, R. J., & Wurtman, J. J. (1986). Carbohydrate craving, obesity and brain serotonin. Appetite, 7, 99–103. Wurtman, R. J., & Wurtman, J. J. (1995). Brain serotonin, carbohydrate craving, obesity and depression. Obesity Research, 3, 477–480.