Elsevier

Appetite

Volume 58, Issue 3, June 2012, Pages 955-963
Appetite

Research report
Use of a clay modeling task to reduce chocolate craving

https://doi.org/10.1016/j.appet.2012.02.044Get rights and content

Abstract

Elaborated Intrusion theory (EI theory; Kavanagh, Andrade, & May, 2005) posits two main cognitive components in craving: associative processes that lead to intrusive thoughts about the craved substance or activity, and elaborative processes supporting mental imagery of the substance or activity. We used a novel visuospatial task to test the hypothesis that visual imagery plays a key role in craving. Experiment 1 showed that spending 10 min constructing shapes from modeling clay (plasticine) reduced participants’ craving for chocolate compared with spending 10 min ‘letting your mind wander’. Increasing the load on verbal working memory using a mental arithmetic task (counting backwards by threes) did not reduce craving further. Experiment 2 compared effects on craving of a simpler verbal task (counting by ones) and clay modeling. Clay modeling reduced overall craving strength and strength of craving imagery, and reduced the frequency of thoughts about chocolate. The results are consistent with EI theory, showing that craving is reduced by loading the visuospatial sketchpad of working memory but not by loading the phonological loop. Clay modeling might be a useful self-help tool to help manage craving for chocolate, snacks and other foods.

Highlights

► Visuospatial interference reduced chocolate craving relative to verbal interference. ► We used a novel visuospatial task, clay modeling. ► Clay modeling reduced craving imagery and intrusive thoughts. ► Results are consistent with Elaborated Intrusion theory.

Introduction

Craving for food is a cognitive-emotional appetitive state distinct from physiological states such as nutritional status (Pelchat & Schaeffer, 2000) or hunger (Lafay et al., 2001), and which plays an important role in the development of obesity (Schlundt, Virts, Sbrocco, & Pope-Cordle, 1993). Craving is associated with binge eating (Gendall, Joyce, Sullivan, & Bulik, 1998) and early dropout from weight-loss programs (Sitton, 1991). Even if it is resisted, craving can be distressing and distracting (Green, Rogers, & Elliman, 2000). Self-report data suggest that the phenomenology of craving for food is similar to craving for addictive drugs, and involves mental images of consumption (Kavanagh et al., 2009, May et al., 2008, May et al., 2004). The present study tests whether a novel technique for blocking imagery can reduce craving for a commonly-craved food, namely chocolate.

A recent theory of the psychological processes involved in craving, the Elaborated Intrusion or EI theory (Kavanagh et al., 2005), gives mental images of consumption a key role in craving. The core of EI theory is the interplay between associative and elaborative cognitive processes. Craving episodes are triggered by environmental and internal cues, including physiological changes, retrieved memories, and associated thoughts and images. Through automatic, associative processes, these cues result in apparently spontaneous intrusive thoughts about desired substances or activities.

Depending on the extent of deficit and whether competing cognitive tasks are present, the intrusive thought may then be elaborated. This cognitive elaboration is the heart of craving, drawing cognitive resources away from other tasks, inducing changes in emotion and enhancing or maintaining motivation to acquire and consume the substance. Elaboration may involve generating expectancies about consummation, or planning how to acquire the substance, but most commonly it involves the construction, maintenance, and manipulation of sensory images of the desired target. For example, someone might imagine selecting a chocolate from a box, taking a bite and feeling the chocolate melt in their mouth, and feeling the pleasure of having their hunger or desire for something sweet satisfied. These sensory images, because they mimic the actual experience, are immediately pleasurable. However, they trigger further substance-related thoughts and increase awareness of the perceived deficit, and so are ultimately aversive, leading the person into a vicious circle of momentarily pleasant imagery that then impairs mood and sucks more resources into creating an even better, more realistic, more pleasurable image, which then increases awareness of deficit and negative affect even further.

There is already evidence for the role of sensory imagery in craving for addictive substances, such as alcohol and nicotine, and in more everyday desires for food, drink, and physical activity. Visual images are commonly reported during episodes of craving for food, drinks, cigarettes, and even during craving for sport (May et al., 2004, May et al., 2008). Sensory imagery predicts the intensity and frequency of craving within and across episodes, and across participants (Kavanagh et al., 2009, May et al., 2008, Statham et al., in preparation, Tiggemann and Kemps, 2005).

Interference with imagery in the laboratory, by the construction of competing but emotionally neutral visual images, reduces craving for chocolate (Kemps & Tiggemann, 2007), food (Harvey et al., 2005, Kemps and Tiggemann, 2007), coffee (Kemps & Tiggemann, 2009), and cigarettes (May et al., 2010, Versland and Rosenberg, 2007). This neutral imagery task has limited use outside the laboratory because participants learn the set of images through repeated use, reducing the task from effortful generation of a novel image to less demanding retrieval of a previous image (Panabokke, 2004).

Craving is also reduced by non-imagery tasks that selectively target appropriate components of working memory. Working memory is the set of mental processes by which we temporarily retain and transform information in memory to allow performance of tasks such as adding up a column of figures, following a conversation, planning a route etc. Temporary retention is achieved by the phonological loop, for auditory and verbal information, and the visuospatial sketchpad for visual and spatial information. These two short-term memory systems are controlled by a central executive, while information from long-term memory is fed into the system, and temporarily stored, by an episodic buffer (Baddeley, 2000). According to EI theory, the visuospatial sketchpad is a key component in craving because it is essential for vivid visual imagery (Baddeley & Andrade, 2000), and the research described above shows visual imagery to be a feature of craving for a range of substances. Imagery of smell, taste and bodily sensations (e.g., swallowing) are also important in food and substance craving, but auditory imagery tends not to be because the activities themselves are not characterized by how they sound (e.g., Kavanagh et al., 2009, May et al., 2008). The visuospatial sketchpad and other components of working memory have limited capacity, therefore loading the visuospatial sketchpad with a task requiring visual or spatial processing prevents that component simultaneously being used to construct vivid visual images (Baddeley & Andrade, 2000). Consistent with EI theory, visuospatial loads have been found to reduce craving. For example, dynamic visual noise (Quinn & McConnell, 1996) reduces craving for chocolate and other foods (Kemps et al., 2005, Kemps et al., 2004, Steel et al., 2006), as do side-to-side eye movements (Kemps et al., 2004). McClelland, Kemps, and Tiggemann (2006) found that the novel task of watching one’s forefinger moving in 1 cm jumps across one’s forehead also reduced food craving. Tasks that load the phonological loop of working memory, such as auditory imagery of sounds, do not reduce substance or food craving (Kemps and Tiggemann, 2007, May et al., 2010), though they may reduce craving for sports (May et al., 2008) and other noisy activities such as racetrack gambling.

If cravings are an important factor in relapse from abstention, and visual imagery is a key component of craving, then these laboratory findings suggest that visuospatial tasks might be useful to help people resist everyday cravings and so support abstention attempts. However, none of the research tasks described above have been tested outside the laboratory and all have limited suitability for take-home use, either because they are too intrusive or embarrassing to perform in public (eye movements, forehead tracking), or because they require custom-made software (dynamic visual noise). Stuart, Holmes, and Brewin (2006) used a novel visuospatial grounding task that potentially solves these problems. The task involved making cubes and pyramids from modeling clay (plasticine), with one’s hands and the clay out of sight. Performing this task while watching a trauma film reduced the incidence of subsequent intrusive images from that part of the film relative to intrusions from parts of the film viewed without a concurrent task, a finding that Stuart et al. attributed to the visuospatial nature of the modeling task. A previous study had shown that a concurrent and demanding verbal task, counting backwards by threes, increased the number of intrusions (Holmes, Brewin, & Hennessy, 2004), suggesting that the effect of clay modeling was via selective competition for visuospatial resources rather than general distraction. We recently showed that clay modeling reduces cigarette craving relative to a verbal control condition (May et al., 2010) but the task has not to our knowledge been tested in other craving domains. The present study tested the effects on chocolate craving of clay modeling compared with verbal interference, with a view to developing a theoretically driven self-help intervention for food craving.

Section snippets

Experiment 1A: Effect of clay modeling and counting backwards by threes on craving

This experiment compared the effects on craving of the clay-modeling task used by Stuart et al. (2006) to reduce trauma imagery. Following Holmes et al. (2004), we chose counting backwards by threes as a comparison task, as well as a do-nothing control condition, to load phonological working memory and to try and match the general resource loads of the clay-modeling task. Craving was induced through abstinence followed by an induction procedure in which participants rated the sensory qualities

Experiment 1B

Experiment 1B estimated the modality-specific and general resource loads imposed by clay modeling and counting backwards by threes by testing their effects on standard tests of visuospatial and verbal short-term memory.

Experiment 2

Experiment 2 improved and extended Experiment 1A in several ways. First, it measured craving using an improved craving scale, the Craving Experience Questionnaire (CEQ). The CEQ allows measurement of craving factors predicted by EI theory, that is, imagery content and intrusiveness, as well as of overall craving strength. These factors emerged as separable components of alcohol craving in a version of the CEQ focusing on alcohol craving (the Alcohol Craving Experience questionnaire, Kavanagh et

General discussion

A clay-modeling task shown to impede visuospatial short-term memory reduced craving for chocolate compared with a simple counting task assumed to load the phonological loop. A more complex counting backwards task, which also impaired visuospatial short-term memory, also reduced craving. This counting backwards task loaded the phonological loop more heavily than did the modeling task, indexed by reductions in performance on a verbal short-term memory task in Experiment 1B. This heavier verbal

References (39)

  • J. Andrade et al.

    Insensitivity of visual short-term memory to irrelevant visual information

    Quarterly Journal of Experimental Psychology

    (2002)
  • A.D. Baddeley et al.

    Working memory and the vividness of imagery

    Journal of Experimental Psychology: General

    (2000)
  • G. Doherty-Sneddon et al.

    Development of gaze aversion as disengagement from visual information

    Developmental Psychology

    (2002)
  • K.A. Gendall et al.

    Food cravers. Characteristics of those who binge

    International Journal of Eating Disorders

    (1998)
  • A.M. Glenberg et al.

    Averting the gaze disengages the environment and facilitates remembering

    Memory & Cognition

    (1998)
  • M.W. Green et al.

    Dietary restraint and addictive behaviors. The generalizability of Tiffany’s Cue Reactivity Model

    International Journal of Eating Disorders

    (2000)
  • K. Harvey et al.

    The nature of imagery processes underlying food cravings

    British Journal of Health Psychology

    (2005)
  • E.A. Holmes et al.

    Trauma films, information processing, and intrusive memory development

    Journal of Experimental Psychology: General

    (2004)
  • D. Kavanagh et al.

    Imaginary relish and exquisite torture. The Elaborated Intrusion theory of desire

    Psychological Review

    (2005)
  • Cited by (0)

    Acknowledgments: Thanks are due to Sarah-Jane Baugh, Keri Davies, Zoe Gofton, Hannah Gooding, Amanda Hicks, Stephanie Pang, Isabel Parsons and Daniel Pirch who worked on this project as part of Plymouth University School of Psychology’s research apprenticeship scheme. Funding for consumables for Experiment 2 was provided under this scheme. Sally Pears was funded by the University of Sheffield’s Women Academic Returnees Programme.

    View full text