Freud theorised that the human psyche was composed of three parts; the super-ego, the ego and the id (Freud & Strachey, 1962). The id was proposed as the instinctual and primitive part of the mind whereas the super-ego operated as the higher moral conscience. Controlling the influences of each on behaviour was the ego. Describing the internal struggle of goals and opposing desires, Freud described the id as being “like a man on horseback, who has to hold in check the superior strength of the horse” (Freud & Strachey, 1962).
Self-control has often been presented as a struggle between dual selves, as Thaler and Shefrin (1981) proposed with their theory of a planner, probably housed in the pre-frontal cortex and the doer, probably housed in the rest of the brain. A dual-self paradigm was also proposed by Fudenberg and Levine (2006), who further proposed that though the long-term oriented “planner” self can exert self-control over the more short-term oriented self, it can be costly and influences such as high cognitive load can inhibit the ability to exert self-control over the short-term oriented self. In a similar conceptualisation of a dual system, (Metcalfe & Mischel, 1999) depicted the two systems as a “cool”, calculating, cognitive system and a “hot” emotional, impulsive system.
Though a single, precise neural location does not seem to account for the entirety of self-control, broader regions and pathways have been highlighted as integral networks for self-control. The prefrontal cortex and frontal subcortical regions have been increasingly implicated in the engagement of self-control (Banks, Eddy, Angstadt, Nathan, & Phan, 2007; Batterink, Yokum, & Stice, 2010; Bechara, 2005; Li & Sinha, 2008; MacDonald, 2008). Rewards for giving in to temptation, whether they are monetary, food, drug or other rewards, activate dopamine receptors in the mesolimbic dopamine system, especially in the nucleus accumbens (Baler & Volkow, 2006). These systems have been shown in neuroimaging studies to become more active when addicts are exposed to images of drug stimuli (Grant et al., 1996) as well as the insula (Naqvi & Bechara, 2009). These neuroimaging results are similar beyond addiction, as shown by studies comparing brain activation in immediate monetary reward versus delayed gratification for long-term monetary reward and finding that mesolimbic activation is associated with choosing immediate gratification rewards (Diekhof & Gruber, 2010; McClure, Laibson, Loewenstein, & Cohen, 2004).
So, with this neuroimaging support for a mesolimbic system being activated in impulsive desire and motivation for immediate rewards, it is interesting to note that, as dual-systems perspectives have suggested, there is a separate neural network responsible for the exertion of executive control in opposition to those desires. When cocaine abusers were instructed to exert conscious control against cravings, heightened activation was measured in the lateral prefrontal cortex (Volkow et al., 2010) even as mesolimbic system activation is inhibited. This pattern of lateral prefrontal activation, and inhibition of mesolimbic regions, especially the ventral striatum, have also been recorded in smoker’s control of tobacco cravings (Kober et al., 2010) and participant delay of gratification for monetary rewards (Delgado, Gillis, & Phelps, 2008).
Self-control has been a long used term in the literature, but it had been little studied as an operationalised construct. In the literature of psychological research it has often been used as a vague descriptive term that captured any part of self, any system, whether neural or otherwise, that described some means of denying gratification, of resisting an impulse to enact some behaviour that was seen as damaging. As such, it was a term easily understood by psychologists and lay people alike, but little examined beyond its descriptive use. For example, an individual may remove all tempting food items from their house, drive a new route home that avoids tempting fast food restaurants and use smaller dishes to serve food on to trick their mind into eating less. This individual might be described as having a high level of self-control. Another individual may leave the tempting food in their house, take the same route home past the fast food restaurants, and serve their food on a regular sized dish but simply eat less. They might achieve this by facing the tempting stimuli directly, but then clenching down, tightening up and resisting those temptations by sheer determination. They too could be described as having a high level of self-control. However, both cases are very different. Though self-control has and continues to be an easily understood, common term to describe an outcome, little had been researched in psychology about exactly what self-control was and by what systems it operates, fails and succeeds.
Baumeister began research which favoured another term for self-control “willpower”. Though this term has been, and is, often used interchangeably with self-control, its use in Baumeister’s research and the very nature of the word itself, hint at a particular theoretical focus. Muraven, Tice, and Baumeister (1998) switch between terms “self-regulation” and “self-control” within the same article, to describe the same construct and yet they also present a case for semantic terminology being important. They reference a traditional concept of “willpower” and argue that it needs to be revived in current research because of the theoretical specifications implied by the particular term (Muraven et al., 1998).
The authors refer to early work by (Thornton, 1939) and (Rethlingshafer, 1942) which used factor analysis to find latent factors among various physical and cognitive tests. Muraven et al. (1998) seem to state that the articles label one large common factor describing the ability to persevere as “willpower”, however the reading by this author finds no such label in the original articles by Thornton (1939) or Rethlingshafer (1942). In (Thornton, 1939) the author concludes that although the tests studied all purported to measure persistence, five separate factors, with some degree of independence, were extracted, with three identified tentatively as describing ability to “withstand discomfort to achieve a goal”, “keeping on at a task” and a kind of gender-related “sex-strength”. In (Rethlingshafer, 1942) three factors are labelled by the author as “the habit of finishing whatever is started”, “endurance” and “strength” and are posited to fall under the umbrella of a larger construct called “persistence”. It is important to note though, that (Thornton, 1939) notes the relationship to “willpower” in the introduction to his article, quoting Allport (1937);
“The concept of persistence is far bolder, and in the long run should prove sounder. If so, it will take care, partially at least, of the thorny problems of “will power” whose inclusions in any psychological schedule of personality is as necessary as it is vexatious.”
This quote illuminates the problem researchers have had in defining self-control, before even attempting to measure, model and explain its possible causes, effects, failures and the strengthening of it. The term “will power” used by Allport (1937) and “willpower” used by Baumeister, Bratslavsky, Muraven, and Tice (1998) describe what is commonly termed self-control or self-regulation, but emphasises the energetic aspects of this control. Freud (1961) highlighted the strength aspect of this control when he described the conflicts between the id and the superego and the struggle of the ego to subdue, control and force the enacting of behaviours. It was for the strength of the ego that Baumeister et al. (1998) termed self-control alternatively “willpower” or “egopower” and the depletion of that theorised power “ego-depletion”. The two aspects implied by the term “willpower”, that a power has a strength and that a power is drawn from some resource, were reflected in two of the main tenets of Baumeister et al. (1998) strength/resource model of self-control. In this model self-control, or willpower, was theorised to rely on a strength, which, like a muscle would have different levels of strength between individuals and could be strengthened or weakened. Also like a muscle, it was theorised to rely on a resource which was limited and could be depleted over time and continued use.
Therefore the term willpower is taken to refer to a more theoretically attached construct, that of the actual strength or resource engaged, exerted and possibly depleted by acts of self-control. The term willpower has traditionally been used by the layperson as a convenient descriptor for an energy that had until recently been dismissed as a construct beyond operationalisation or ignored by psychological research. Though self-regulation and self-control research have been prevalent throughout the literature over time, Baumeister’s research has sparked renewed interest in an operationalisable construct of the actual power, resource or energetics of willpower.
The strength/resource theory of willpower generated much renewed research interest in self-control with research supporting its hypotheses and other research challenging its hypotheses. Because of the theoretical implications implied by use of the term “willpower” many researchers use either self-control or self-regulation to describe the construct. Often researchers will use both interchangeably within an article, or with unclear differences within the title of a paper, such as “ ‘Willpower’ over the life span: decomposing self-regulation” (W. Mischel et al., 2010), “Affective forecasting and self-control: Why anticipating pride wins over anticipating shame in a self-regulation context” (Patrick, Chun, & MacInnis, 2009), “Making choices impairs subsequent self-control: A limited resources account of decision making, self-regulation, and active initiative” (Vohs et al., 2014) and “Self-regulation and limited resources: Does self-control resemble a muscle?” (Muraven & Baumeister, 2000).
Although the terms are even more frequently used and used interchangeably than willpower, self-control and self-regulation do imply different constructs. Control evokes images of a harder, more forceful clamping down of, pushing back against or resisting of some powerful urge or impulse. Regulation evokes a more subdued, regular and consistent maintenance, refinement and updating.
Kuhl and Fuhrmann (1998) theory of volition postulates that self-regulation and self-control are separable elements of volition. Self-regulation is defined as the maintenance of one’s actions to align with one’s self-image, whereas self-control is defined as controlling actions to support pursuit of an active goal. Though subsequent research has shown that according to some measures, there is overlap, or shared variance between the two constructs (Hanfstingl, Andreitz, Müller, & Thomas, 2010), research following Kuhl and Fuhrmann (1998) theory of volition continues to define and measure self-regulation and self-control as distinct constructs (Dewitte & Lens, 1999; Forstmeier & Rüddel, 2008; Hanfstingl et al., 2010; Wojdylo, Baumann, & Kuhl, 2017).
Recently other researchers have recognised the need for clarification between the terms self-regulation and self-control, with one distinction given being that self-regulation is a larger, more broadly encompassing construct that can include self-control in some situations (Berkman, Kahn, & Livingston, 2016; Fujita, Carnevale, & Trope, 2016; Kotabe & Hofmann, 2015).
Mann, De Ridder, and Fujita (2013) describe self-regulation as a long-term process that involves the employment of various strategies in order to decide on goals, continually pursue and monitor progress toward goals and to maintain various aspects of self, such as physical and emotional wellbeing. As such it is described as the employment of various strategies in order to behave according to some self-set standard. Fujita et al. (2016) explain that self-control is a type of self-regulation, employed for successful self-regulation, but that not all self-regulation involves self-control. For example, on a physical level, Fujita et al. (2016) give a basketball player taking a shot as an example. The player must co-ordinate and adjust their feelings, focus and each movement of the body so that they achieve their desired standard of successful behaviour. However, there is no conflict between desire and goal, no impulse to control and so this is an example of self-regulation, but not self-control.
Generally, a self-control situation involves a dual-motivation conflict in the present, between a higher order goal and a tempting desire (Fujita et al., 2016). Self-control is often described as a conflict between enacting or resisting behaviour in service of a more long-term, distal outcome or enacting behaviour that gives immediate gratification or reward (Metcalfe & Mischel, 1999; W. Mischel, Shoda, & Rodriguez, 1989). The competing desires and goals are often conceptualised as “cool/cold” inputs, such as considerations of long-term finances or goals for health or losing weight, in opposition to “hot” inputs, such as the impulse to gamble, or the craving for some tasty, but goal-incongruent food (Berkman et al., 2016; Metcalfe & Mischel, 1999; W. Mischel et al., 1989).
With these distinctions in mind, it is clear that self-regulation is a larger, broader and more consistent, regular construct that may employ self-control at times but does not necessitate it. For example, an individual with a goal to lower body fat may be fortunate enough, or have enough foresight, to employ various self-regulatory strategies, such as planning, avoidance of temptations, motivational strategies and more, to never feel any large degree of temptation or conflicting desire and thus avoid ever needing to use self-control at all. In the emotional realm, one’s personal standard of self-regulation may be to remain calm and not express any stress. An individual may spend a week, with their schedule, planning, avoidance of certain people and situations allowing them to remain calm. This week, having been self-regulated might entirely avoid the use of self-control. However, one day they may be unexpectedly stuck in traffic, late for a frustrating meeting with frustrating people and may have to exert self-control to calm their building stress and to prevent their frustration from showing on their face and in their reactions.
So, self-regulation is a broader, more general term that might involve the employment of various strategies, including self-control, to maintain a personal standard or to work progressively toward a new standard, in the form of some goal. Self-regulation involves volitional and conscious use of executive functions for planning, approach or avoidance of environments, deployment of focus and attention and cognition in ways that will support the maintenance or pursuit of a standard or goal.
Self-control likewise involves the exertion of executive function but in a more effortful manner, initiated by a present conflict between two (or more) competing motivations, a goal-directed, distal outcome driven motivation, or a goal-incongruent, impulsive, immediate or proximal gratification or reward based motivation.
Just as self-regulation might include, but is not limited to, self-control, so too does self-control comprise different aspects. One of these is inhibitory control. Inhibitory control describes a self-control process whereby some impulse, whether physical, emotional or cognitive is resisted, controlled or inhibited, an avoidance action. These inhibitory control actions are those typically associated with self-control, resisting the impulse to eat a second piece of cake, stopping one’s self from automatically lighting up that cigarette, resisting the impulse to look at some sexual sight. However, self-control may also involve approach actions, pushing behaviour toward some response rather than pulling away, pushing a certain behaviour rather than some alternative, or rather than a more comfortable inertia. Apart from the proactive actions, or inhibitory control, self-control may also entail the effortful control of attention or focus, as well as motivation to enact a target behaviour.
Though some researchers also use the terms self-control and inhibitory control interchangeably, for example (Tabibnia et al., 2011). So although they are separable terms, inhibitory control, both in the public image and researcher focus, does comprise a large portion of self-control (Morasch & Bell, 2011).
A similar term that is used less often in research, but in the public parlance, is “self-discipline”. This term refers not to a state or process, such as self-regulation or self-control or inhibitory control, but to a consistent, stable, enduring trait. This trait may be composed of self-regulation and self-control elements and is measured by subscales of personality tests such as the NEO tests (McCrae, Costa, & Martin, 2005) and the Volitional Components Inventory (Kuhl & Fuhrmann, 1998).
However, there is also a test born of the research by Baumeister which raises the issue of blurred lines between these constructs. The Self-Control Scale and the Brief Self-Control Scale (Tangney, Baumeister, & Boone, 2004) have been widely used in research to measure a trait level of self-control, however recently some research has raised questions over what the scale measures. The suggestion that self-control is a trait, as stated in the scale name, is indicative of the ambiguity in terminology in the field. Based on the definitions of each term given here, the long-term nature of each construct, ranked in descending order from more stable, enduring trait level to more fluctuating state level would be; self-discipline, self-regulation, self-control (and inhibitory control and willpower). Based on these definitions, the (Brief) Self-Control Scale perhaps measures self-discipline or self-regulation, rather than more state based constructs such as self-control, although it could be said that the trait version of self-control, the accumulated total of regular self-control instances is self-discipline or self-regulation by definition.
An experience sampling research design Hofmann, Baumeister, Förster, and Vohs (2012) showed that the Self-Control Scale may not be measuring what it was assumed to. When participants in the study were beeped at random intervals each day, for a week, they were required to detail their experiences of temptations, desires and self-control through the day. The assumption was that those who scored highly on the trait Self-Control Scale would show stronger self-control in the face of temptation than those low on the scale scores. However, it was found that scores on the Self-Control Scale were negatively associated with the exertion of self-control in the participant’s daily lives, participant’s that scored more highly on the Self-Control Scale exerted less resistance, or control effort, over the course of the week (Hofmann et al., 2012). The authors concluded that the Self-Control Scale may not measure a greater strength and more regular successful exertion of self-control moment by moment, but may measure patterns, habits and routines that have been built to avoid temptation rather than facing it. This explanation was supported by a negative association of scores on the Self-Control Scale with the strength of tempting desires reported by participants, the higher an individual’s Self-Control Scale score, the less tempting desire they experienced.
If this is the case, that the Self-Control Scale measures an individual’s ability to construct their routines, habits and environment in such a way as to avoid temptation rather than fight against it, it better fits the definition of self-regulation offered above.
One potential cause of confounding results in use of the Self-Control Scale may be its application in research as a unidimensional measure, when research suggests it may be a two dimensional measure (Maloney, Grawitch, & Barber, 2012). A factor analysis was conducted by Maloney et al. (2012) and suggested two factors; restraint and impulsivity. Though these two factors appear similar, or to be opposite ends of the one continuum, with an assumption that if one is high in self-control then they are low in impulsivity and if high in impulsivity then they are low in self-control, however research suggests they are slightly different. Freud described the id, giving rise to impulses, as a separate force, interacting dynamically with the ego, which exerts control over these impulses (Freud, 1961). Rather than being descriptors given to two ends of the same spectrum, they were theorised as separate forces.
Carver (2005) has surveyed literature from varied fields and theoretical approaches and describes links between three and five personality models and the constructs of constraint/restraint and impulsivity, describing elements of sensation seeking and a need for exciting experiences or arousal. Biological approaches suggest that a Behavioral Approach System (BAS), encourages approach behaviours, motivating action toward reward, whereas a Behavioral Inhibition System (BIS), encourages avoidance behaviour, withdrawing or resisting behaviour with potential negative consequences, and that the balance of these could be equated with impulsivity levels (Carver, 2005). Cognitive models have described two mental systems that can operate independently and simultaneously to combine in determining resultant behaviour. These can be described as the experiential/emotional/impulsive/hot system and the rational/deliberative/restraint/cool system (Carver, 2005; Metcalfe & Mischel, 1999). This distinction between impulsivity and restraint is further clarified by developmental theories by (Eisenberg et al., 2004), whose research suggests that both reactive undercontrol (equivalent to the Behavioral Approach System) and reactive overcontrol (equivalent to the Behavioral Inhibition System of avoidance) motivations are impulse driven, whereas a third factor of effortful control (restraint) emerges later in development.
Theories of a dual mode of operation have much support from varied realms of psychology, biological, social, developmental, personality and cognitive, but are not without challenges (Carver, 2005). It is generally theorised in dual-mode models that these two modes of mental processing can operate simultaneously and sometimes in conflict, both hot and cold, reflective and reactive. Neural imaging does seem to support these two-mode theories, with deliberative, reflective control processes having been shown to engage the anterior cingulate, prefrontal cortex and hippocampus while automatic, reactive, impulsive processes appear to engage the lateral temporal cortex, amygdala, and basal ganglia (Casey, Tottenham, & Fossella, 2002; Lieberman, Gaunt, Gilbert, & Trope, 2002). Lieberman et al. (2002) extended the theory to posit that behaviours are not permanently restricted to one mode or the other, but that any behaviour repeated enough could pass from processing in the reflective, deliberative mode to the reactive, automatic mode.
These dual systems theories are supported by (Hofmann, Friese, & Strack, 2009) who also offer a helpful definition of impulses. Hofmann et al. (2009) distinguish impulses from drives, or more global desires by describing an impulse as being specific, occurring when global motivations align with specific stimuli in the environment. So, though hunger is a drive, or global motivation, it might trigger an impulse when paired with the presence of cake on a nearby table. Hofmann et al. (2009) also define an impulse as having a strong hedonic reward or incentive, being spatially or temporally immediate (immediately available for gratification), and as carrying a strong inclination to perform a certain behaviour.
From this definition of an impulse it is easier to define self-control as a separate, though related, construct to self-regulation. Although self-regulation can occur without the presence of impulses, self-control arises as an effortful executive decision to inhibit or resist an impulse, or to enact a behaviour despite an impulse to do otherwise (Hofmann et al., 2009) (Metcalfe & Mischel, 1999; Tangney et al., 2004).
Before psychology as a field existed, Greek philosophers Aristotle and Socrates argued over the issue of passion versus reason. Socrates claimed that it was not a weak will that failed, but faulty reasoning, that some mistake or lack of foresight in reasoning led people to act for the short-term rewards rather than the long-term. Aristotle, on the other hand, claimed that people sometimes acted against their better judgement, because they were overpowered by their passion. Religion has traditionally viewed self-control as a virtuous struggle to impose a reasoned, better judgement on behaviour, rather than being overcome by passions, or impulsive temptations, as seen in the works of Saint Augustine and others (Geyer & Baumeister, 2005). The book of Proverbs in the Bible shows the level of virtue afforded to self-control. Proverbs 16:32 “Better a patient person than a warrior, one with self-control than one who takes a city.” And Proverbs 25:28 “Like a city whose walls are broken through is a person who lacks self-control”.
Early psychologists such as Freud and William James described an inner struggle, and balancing challenge between forces, id, ego and super ego with Freud, or impulsive forces and will with James. More recent psychology has seen study on the delay of gratification and the monitoring of internal states and judgement of alignment with standards, along with the resource depletion and failure of self-control. (Strack & Deutsch, 2004) proposed a dual-systems perspective called the Reflective Impulsive Model (RIM) and other theories have posited similar dual systems perspectives using different terminology, but all proposing that there are fundamentally two different mental systems that process differently and rely on separable neural areas.
Impulses are assumed to arise from a neural activation of associated clusters (Metcalfe & Mischel, 1999; Strack & Deutsch, 2004) that have been either created (hereditable, genetic, foundational circuits) or built and strengthened by experiences that have paired together spatially or temporally to create strong neural pathways and strong experiential associative reactions to stimuli. For example, an individual may link multiple stimuli, hunger, the sight of cake, the concept of cake, the positive feelings (both physical and emotional) generated by the consumption of cake, and the behaviour of putting the cake in one’s mouth. This associative cluster may be triggered by the drive state of hunger paired with the stimulus of a cake being present, or it may be triggered by the concept of cake alone, or by other drive states such as negative affect. These associative clusters can be very efficient ways for an individual to enact behaviour to control various states of dysregulation such as hunger, thirst or negative affect. They are usually built and strengthened slowly over time and require no attentional resources, executive function and do not require the individual to reflectively approve or disapprove of the impulse and resultant behaviour, in fact impulses can often occur without the individual’s conscious knowledge (Strack & Deutsch, 2004).
The reflective system, on the other hand, operates at a conscious level and involves slower, more deliberate processes including executive functions such as judgements, accessing working memory, focus, targeted deployment of attention, evaluations and more. Various accounts of a dual-process theory posit that these higher order, reflective and executive functions are limited, whether by resources of simultaneity in time (only so many executive functions can be employed effectively at the same time) or by a physical fuel that can be depleted (Evans, 2008; Hamilton, Vohs, Sellier, & Meyvis, 2011; Muraven & Baumeister, 2000). These resources may affect a self-control situation at various points, for example by reducing the ability to monitor the degree to which current state matches standards, or by reducing the amount of self-control exerted on a problem if it is registered.
As Strack and Deutsch (2004) describe, when an impulse is incongruent with the reflective system’s higher order goals, or standards for self, then various behavioural schema may be activated in the motor cortex, priming various actions that may be enacted, some aligning with impulses and some aligning with reflective processes. How it is decided which of the activated behavioural schemas is enacted is the topic of much debate and much that is as yet unknown. Various precursors and situational conditions will add or reduce weight given to reflective or impulsive processes. Various constraints and conditions leading to self-control failure have received much research over the last two decades, but relatively little has been given to studying the influences that are precursors for impulses and this may be a fruitful focus for successful interventions.
There is a strong body of research into the dual process model, with restraint and impulsivity not being merely opposite ends of the same spectrum, but separate types of mental processing with distinct neural substrates. Maloney et al. (2012) confirmed the two-factor structure of the Self-Control Scale, with the factors named restraint and impulsivity. They argue that a lack of discrimination between the two related, yet distinct constructs can lead to a misuse of the scale. In their analyses they state that the restraint factor is synonymous with a tendency for self-control, but also note that there are items which may be referring to self-discipline. The four items which comprise the impulsivity factor refer to a dispositional tendency to act spontaneously on thoughts or feelings. Other conceptualisations of the Brief Self-Control Scale as a 2 dimensional model have shown similar constructs with some item overlap (De Ridder, De Boer, Lugtig, Bakker, & van Hooft, 2011; Ferrari, Stevens, & Jason, 2009).
A recent study by (Lindner, Nagy, & Retelsdorf, 2015) used “satisfaction” as an outcome measure, using the Satisfaction With Life Scale to test the predictive power of the Brief Self-Control Scale. In the article running confirmatory factor analysis comparing the goodness of fit and the correlation with outcome measures of a unidimensional Brief Self-Control Scale and three versions of a two dimensional model it was found that the unidimensional model, taking the total BSCS scores gave the best goodness of fit and that two dimensional models did not add significant incremental predictive validity beyond that.
Therefore, despite questions about the dimensionality of the Brief Self-Control Scale, the scale does have significant research support for its predictive validity. Duckworth and Kern (2011) conducted a meta-analysis of self-control measures and note that although self-control seems to be a multi-dimensional construct, for which various measure types should be used in combination, that amongst them self-report surveys may be the best in terms of variance explained and ease of administration.
(Whiteside & Lynam, 2001) created a four factor measure for impulsivity, titled the UPPS, standing for the four factors of urgency, (lack of) premeditation, (lack of) perseverance, and sensation seeking.
(Hofmann et al., 2009), in writing about impulsivity and restraint, raise an important point about the research into self-control, noting that there has been much research into the failure of self-control ability or strength or resources, but relatively little into the potential other cause of self-control failure, the level of impulse or desire. Within current dual systems perspectives, there are two reasons why self-control may fail, one is that the individual’s self-control resource is too low/insufficient/weak and the other is that the gap, or conflict, or incongruence between their impulsive, tempting desire and their goal or standard is too great. On this point there are two factors that may make the conflict too great for self-control to overcome, one is that their goal or standard is too high, as may be the case with perfectionists, the other is that their impulse or desire is too strong, as may be the case with addicts.
So, when an individual has a goal to avoid eating any chocolate but succumb and eat chocolate there are two possibilities. The first is that, even though the desire they felt was not so strong, they were tired, stressed, distracted, or for some other reason their current levels of self-control strength were low, and thus insufficient to overcome even a weak desire. The other is that, even though they were refreshed and had full use of their self-control strength, the level of conflict they felt (defined as the gap, or degree of incongruence between their higher order goal and their tempting, impulsive desire) was too large, even for their high level of self-control. This particular failure may have two causes, their impulsive desire is so strong, that no matter the level of their higher order goal, the gap from one target (immediate gratification in the form of chocolate, a desire strongly felt and often felt) to the other (the higher order goal of abstinence from chocolate) is too large. The other cause may be that, although their desire is not so strong (they feel like eating chocolate a little, from time to time) the gap between it and their higher order goal is large because of very high standards (they target complete abstinence from chocolate at all times). It is worth noting that, although these three aspects of a self-control situation (self-control is too low, desire is too high, goals are too high) are distinct inputs into the model, they are relative terms. A low desire may be low enough in one situation when self-control is high, but the same level of desire may be too high at a time when self-control is relatively low.
In any case though, the end result is the same, either a subjective experience of self-control success or self-control failure. Regardless of which of the three inputs is to blame, the individual may think of the failure as being due to a lack of self-control strength, when in fact it might best be explained by an impulsive desire that is too great. All this is to say that, there are in fact two targets for intervention to improve self-control success; to raise self-control strength, or to decrease the gap, or level of conflict or incongruence, between impulsive desire and higher order goal. This second target has within it two approaches, one is to decrease/increase/align impulsive desires with goals, and the other is to increase/decrease/align goals to align with impulsive desires.
With this lack of focus in the research on the adjustment of impulsive desire for better self-control, it is interesting to note the theoretical approach and explanation taken by the researchers for the full (and Brief) Self-Control Scale (Ent, Baumeister, & Tice, 2015). They describe studies which show that individuals high in “trait self-control”, as measured by the Brief Self-Control Scale, also reported more frequent avoidance of temptations, avoidance of distractions and choosing friends who seemed likely to help them achieve their goals. Individuals with higher “trait self-control” also exhibited an increased tendency to choose distraction free environments to perform laboratory tasks in the experimental research, even though it meant a less exciting version of the task, or a longer wait to begin (Ent et al., 2015). Notable in this research paper are a few things. First of all, it supports the conclusions indicated by earlier experience sampling research (Hofmann et al., 2012), that suggested that those measuring high on the Brief Self-Control Scale were not more regularly better at actively resisting temptations than their low “trait self-control” counterparts, but were actually proactively enacting routines and behaviours that avoided temptation and distraction. Secondly, the research article of Ent et al. (2015) evinces the confusion in terminology use, even as they attempt to clarify it;
“Attempting to resist impulses as they arise (rather than avoiding such impulses) may be a relatively ineffective self-regulatory strategy. The capacity to resist impulses depends on self-regulatory capacity generally. Recent work has suggested that each person’s capacity for self-regulation fluctuates across time, presumably because each act expends and depletes a limited resource, so that one’s willpower occasionally becomes depleted (Baumeister, Vohs, & Tice, 2007; Hagger, Wood, Stiff, & Chatzisarantis, 2010). Hence, if people rely solely on their willpower to resist temptation, they are likely to fail periodically, if only because some temptations will be encountered when one’s powers of resistance are low.
Effective self-control might therefore involve more than resisting temptation. (Fujita, 2011) made a persuasive case for broadening the focus of self-regulation research beyond effortful inhibition of impulses. Although resisting temptation and inhibiting problem desires are undoubtedly useful capabilities, avoiding tempting situations so as minimize problematic impulses could reduce the risk of self-regulatory failure.”
This paragraph highlights confusion in the terminology and exactly what is being measured by the Brief Self-Control Scale. The first sentence highlights that self-regulation encapsulates various strategies and underlying constructs, including self-control (or the effortful resistance of impulses). These definitions are in accord with those offered previously in this current piece of writing. The following two sentences, however, introduce considerable confusion to the definition of constructs by suggesting that self-control or inhibitory control capacity depend on self-regulatory capacity. Are they suggesting here that self-regulation and self-control are the same thing and that this single self-regulation/self-control construct has a limited capacity, reflected in the limited capacity to resist impulses (among other things)? Or are they suggesting that self-control (or resistance of impulses) depends on a limited capacity, just as self-regulation also depends on a limited capacity? Based on the definitions of these constructs offered herein, self-control (as a situational effortful exertion) may be a limited resource (as suggested by ego-depletion research), but self-regulation does not draw from a limited resource (as far as the research literature has thus far discovered). Self-regulation, as defined herein, is only limited in capacity by one of the tools under its employ: self-control.
The third sentence in the above paragraph becomes even more problematic, using the terms self-regulation and willpower. It is suggested that the term self-regulation actually refers to self-control, so the sentence should read “each person’s capacity for self-control fluctuates over time”. The following use of the term “willpower” is one that is theoretically tied to the idea of “ego-depletion”, the theory that self-control is like a muscle with a capacity dependent on two things; the strength and the resources/fuel available at any given time. Therefore self-control refers to the act, whereas willpower refers to the fuel or energy actually being used in that act. It should be noted that willpower, tied as it is to the theory of ego-depletion, is a somewhat controversial term. Though the author thinks it likely, based on the current body of research, that self-control may rely on some kind of limited resource, and be restricted by the strength capacity of each individual, it is unclear whether this occurs in a manner exactly as described in the strength/resource model. The research on the strength/resource model is far from equivocally proven in the current body of research and thus caution is advised in using more theoretical terms such as “willpower” or “ego-depletion”, as they presuppose conclusions that are not yet conclusively proven in the research.
The first sentence of the second paragraph again highlights the confusion as they seem to use the terms self-control and self-regulation interchangeably, which, again, in this author’s opinion should be kept as distinct constructs, as defined above, for clarity of research goals and the dissemination of research. These paragraphs do highlight the struggle for expanded understanding of these constructs however, along with (Fujita, 2011) noting that self-control can be more than impulse control, or inhibitory control. Self-control may also entail proactive action, rather than inertia or an alternative action. (Fujita, 2011) defines self-control as involving a dual-motive conflict, one a proximal, or immediate reward, versus a distal, or long-term reward. Thus eating the cake involves a motivation for immediate reward, while abstaining in hopes of better health in the long-term, or future weight-loss would be a behaviour motivated by the distal reward. It is worth noting that (Fujita, 2011) also notes this as a distinction between self-regulation and self-control. (Fujita, 2011) Describes how self-regulation of a basketball shot involves the monitoring, coordination and focus on arm movements and targeting and thus may be described as self-regulation, but not as self-control, as it does not involve a dual-motive conflict.
This definition of self-control as involving a conflict is supported by (Schmeichel & Zell, 2007) as they note that it is the “attempt to override or alter one’s dominant response tendencies”. The article describes stable differences in the capacity of individuals to exert this effortful control, as measured by the Self-Control Scale, a view that is in contrast with recent research by and suggesting that it may measure individual’s ability to avoid temptation, rather than overcome it. However, early experiments by (W. Mischel et al., 1989) and later follow-up research (Eigsti et al., 2006) suggested that as early as 4 years old, individual differences in the ability to delay gratification were evident, exerting self-control to resist an immediate reward in favour of a greater long-term reward. It could be said that the techniques employed by these children involved the redeployment of their attention, an effort to avoid or diminish the temptation felt by the present rewards. This explanation aligns somewhat with the recent explanation given for why those high in “trait self-control” as measured by the Self-Control Scale, seem to exert self-control less, by avoiding tempting situations and distractions. However, other research on self-control in children (Kieras, Tobin, Graziano, & Rothbart, 2005) showed that children who exhibited higher inhibitory control (specifically the ability to slow down or suppress motor responses) exhibited better socialisation in emotion suppression when receiving unwanted gifts.
A further study showed that self-control in persisting on a frustrating and difficult motor control task (Brandon et al., 2003) also predicted better smoking cessation in individuals. What these two studies, and perhaps Mischel’s study of delayed gratification show, is that using various measures of self-control capacity, including inhibitory control, individual differences can be seen and predict other positive life outcomes. However, this capacity for greater situational exertion of self-control does not seem to be exactly what the Self-Control Scale is measuring. Rather, it seems to be measuring successful self-regulation strategies that avoid temptation. Perhaps these routines and strategic behaviours may have required initial expenditure to start and then to maintain for long enough to become more habitual, but by the definitions given herein, new research shows that the scale may not measure self-control directly, but measures broader self-regulation strategies, or a trait perhaps better described as self-discipline, in order to avoid having to exert self-control. This does not preclude the scale from possibly also measuring self-control, but thus far there is little evidence that it measures self-control directly.
A further two experimental research designs focussed specifically on the predictive value of the Self-Control Scale for other directly observed self-control behaviours (Schmeichel & Zell, 2007). One study found that high scores on the Self-Control Scale predicted increased inhibitory control in resisting the homeostatic urge to blink compared to individuals with lower scores on the Self-Control Scale (Schmeichel & Zell, 2007). A second study showed that high scores on the scale also predicted longer times for pain tolerance in a cold pressor persistence task, with high scoring participants holding their hand in cold water for a statistically significant amount of extra time (Schmeichel & Zell, 2007). As the authors (Schmeichel & Zell, 2007) note, these findings are interesting because the Self-Control scale does not directly refer to any motor inhibition skills in any of the items. This suggests that a common self-control related construct may exist and contribute to both motor inhibitory as well as broader self-control related tasks in life.
As (Schmeichel & Zell, 2007) note, their research also shows that people can be accurate to some extent when self-reporting their own self-control, as their self-report survey (the Self-control Scale) predicted actual inhibitory control performance and persistence on difficult or painful tasks. The authors also controlled for mood and arousal, but the relationship between the self-report of self-control and the behavioural success of self-control remained statistically strong regardless of shifts in state affect. The authors also note that self-control may be a causal factor in building and maintaining high self-esteem.
Although the authors (Tangney et al., 2004) originally presented five factors within the Self-Control Scale (In descending order of variance accounted for (from 10.2%): Capacity for Self-Discipline, Tendency to deliberate action, healthy habits, self-regulation in service of work ethic, and Reliability) they gave little explanation in the published article and explained that because correlations between factors were high they focus on total scores, suggesting it as almost a unidimensional construct. Also of note is their use of the term “self-discipline” for the first factor which accounts for the most variance. This terminology aligns with the definitions offered herein and in fact, based on more recent findings of the scale seeming to measure broader self-regulatory habits, routines and behaviours beyond strictly defined self-control, self-discipline may be a better term used to describe the scale and its measured construct.
As discussed, a recent analysis of multi-,two- and uni-dimensional interpretations of the Self-Control Scale suggested that a two-dimensional interpretation offered slight incremental predictive validity (conceptualised as inhibitory self-control and an initiatory self-control factors) over a unidimensional interpretation, but concluded that overall it is generally best to use total scores as a unidimensional model in also considering parsimony and ease of use. The initiatory and inhibitory interpretations of the two factors, suggested by (De Ridder et al., 2011) make some sense in explaining the results of recent research on the predictive power of the Self-Control Scale. Earlier research had shown that high scores on the scale predicted more successful motor inhibition. Recent research suggested that those with high scores on the scale exerted effortful self-control to resist temptation less, but had successful self-discipline, or self-regulation, by the maintenance of routines, behaviours that allowed them to avoid temptation and distraction. As noted, these may have required initial exertion of both inhibitory and initiatory self-control exertion to start, but perhaps lean more on initiatory self-control and other, non-self-control self-regulatory behaviours to maintain.
Therefore it is recommended here that, until new evidence suggests otherwise, that the Self-Control Scale rather be thought of as measuring self-discipline (essentially equivalent to accumulated, regular trait levels of self-control exertion) rather than the confusing moniker of “trait self-control”. Furthermore, it is suggested that although the scale may generally be taken as a unidimensional measure of overall self-discipline (or regular, successful use of self-control and other self-regulatory behaviours), that the potential explanatory power of two underlying dimensions to self-discipline (initiatory, or proactive behaviours that may help in avoiding temptation or distraction as well as inhibitory ability, which may not be needed as much when the first dimension is successfully employed) should not be forgotten.
– (Rothbart, Ahadi, & Evans, 2000) calls it “effortful-control” and describes it as being based in anterior cingulate cortex, which is involved in monitoring and planning (among other neural regions, such as parts of motor control) and being an attentional system, responsible for monitoring, and when an error in behaviour is found and then in inhibiting a dominant (stronger, automatic) response, in favour of a subdominant response. The fact that there are such rich attentional aspects is reflected in the simpler self-regulation strategies of babies, infants and toddlers, where simply shifting attention can be a powerful way to regulate emotionality.
WHere DoeS Self-Control Come From?
Before we think about improving self-control capacity, or avoiding situations where failure seems inevitable, it is helpful to first consider one’s current level of self-discipline and self-control capacity. Why is it different between individuals? How does one arrive at their current state? Is it genetic, learnt, or otherwise influenced by environmental factors? It should come as no surprise that it is a combination of various factors.
Research has shown that some variance in impulsivity and also sensitivity to developing addictive behaviours is explained by genetic inheritance (Kreek, Nielsen, Butelman, & LaForge, 2005). Prolonged exposure to addictive substances can also cause significant and prolonged neural changes. For example, prolonged exposure to drugs of abuse can affect neural networks, neural sensitivities to reward and thus drives, expression of genes and more (Kreek et al., 2005).
(Feldman, Greenbaum, & Yirmiya, 1999)
(Pratt, Turner, & Piquero, 2004)
(Sethi, Mischel, Aber, Shoda, & Rodriguez, 2000)
(Romer, Duckworth, Sznitman, & Park, 2010)
(Vazsonyi, Roberts, Huang, & Vaughn, 2015)
So it appears that there are developmental stages for the construction of cognitive abstraction, attentional deployment and other executive functions necessary for self-control. But what is it that differentiates one individual from another? As mentioned, genetic inheritance undoubtedly plays a role in impulsivity and sensitivity to certain temptations, or to immediate rewards, but what shapes how self-control is learnt differentially? One important factor is the family and parental bonds.
(Feldman et al., 1999) describe the mother-infant interactions and test theories in their research. Face-to-face interactions between mother and child start at around 2 months old and involve a great deal of social and cognitive information. The split-second lagged synchronising of facial expressions, or affective expressions between mother and child have been theorised to lead to the smooth transition from mutual regulation to self-regulation. These interactions are believed to be important because they give the infant practise at interpersonal communications, the mutual regulation of positive arousal and also begin building the lead-lag pattern of adult communications in social self-regulation.
(Kopp, 1982) defines some integral steps of the development of self-regulation in infants, from “co-ordinated interactions” at 3 months, and to self-control by the age of 2 years, which involves a kind of internalised socialisation, which include a child’s ability to follow parental commands and the ability to delay behaviour or gratification on request, behavioural dos and don’ts that form the basis of an emergent morality. (Kopp, 1982) conducted longitudinal research on mother-child dyads, measuring synchrony, temperament and other constructs at n and n and n, and found that both temperament (likely a biologically inherited trait) and synchrony predicted the level of self-control a child would develop at 2 years old. (Kopp, 1982) summarises literature detailing developmental stages of this development, stating that during their second year children increasingly show behaviours indicating selfhood and autonomy which leads them to monitor their own behaviour in comparison to separate external objects or individuals and adjust accordingly: the emergent steps of self-regulation which has been described as one of the central cognitive-development milestones.
In the first 3 months of an infant’s development, very simple biological control emerges in the form of control of state of arousal, and the extent of this development is moderated by the extent of neurophysiological development, parental interactions and routines such as feeding or sleeping times. At 9-12 months, an early precursor to self-control emerges with compliance and responsiveness to warning signals and the extent of this development is also mediated on the quality of maternal interactions as well as a latent bias toward social behaviour in general. In the second year, between ages 1 and 2 years old, impulse control emerges, with a greater balance between action and speaking, and this is mediated by the emergence of language skills, the ability to reduce tension, and the sensitivity of parents to the child’s needs. Initial self-regulation emerges next, as the child internalises behavioural patterns for better social conduct as well as motor inhibition. This continues the reaction to external situational cues, including adult commands, from the second year, but builds to self-regulation and auto-regulation of speech and behaviours through years 3,4 and 5. This stage is greatly mediated by social interactions and growth of deeper language learning and deeper understanding.
From 5 or 6 years old onwards, children generally develop into a broader, more generalised self-regulation with adoption of rules that govern general behaviour and speech, irrespective of situational pressures, something akin to morals. This stage is mediated by the developmental extent of cognitive processes such as abstraction, attentional deployment, plans and the ability to think more of the future.
These stages were later verified in research by (H. N. Mischel & Mischel, 1987) who discuss their research into the childhood development of delay of gratification strategies, a facet, or specific application of self-control. (H. N. Mischel & Mischel, 1987) notes that the ability to delay gratification is one of the core facets of ego development in many personality theories. They observed in experiments that children employed attentional or cognitive strategies to successfully delay gratification, either distracting themselves away from the temptation of the object of immediate gratification, or removing it from sight, or by focussing on other things, thinking about the greater, long-term reward. These strategies rely on other executive functions and were observed to be developed in distinct stages. For example children aged 3-5 generally did not conceive of or use attentional deployment, distraction or cognitive strategies to delay gratification, whereas children aged 8-9 appreciated and generally began to use such strategies, while children aged 11-12 systematically appreciated and employed these strategies. In a second study (H. N. Mischel & Mischel, 1987), the intermediate ages of 5 and then 6 year olds saw a steady progression in the increased use of strategies to avoid attention and focus on the tempting immediate rewards. Based on interviews with these participants, it appears that the cognitive development of children at around age 4 does not allow them to separate what they want and what will help them get more of it. Thus the concept that sometimes one must avoid attention or temptation toward what one wants in order to better facilitate self-control and delay of gratification does not seem to emerge as a cognitive ability until later in development.
In earlier research (W. Mischel, 1958) had found that cultural background and absence of a father figure in the household also contributed to variance in children’s ability or desire to delay gratification by refusing an immediate smaller reward in favour of a later but larger reward. It should be noted that these findings are suggestive rather than conclusive, due to the small sample size studied and the correlational nature of the research. Also, the explanation for why absence of a father figure in the household might result in a lower desire or ability to wait for a larger reward is not clear. Perhaps the fathers that left their families passed on a genetic propensity for immediate gratification and lesser ability to endure for later rewards to their children. Perhaps the absence of the father resulted in a lack of good teaching or role-model for self-regulation strategies. Perhaps the absence of a father figure resulted in economic pressure on a single parent family and thus engendered an environment where any food, money or reward would be taken as soon as available, out of necessity rather than choice. This last explanation was controlled for by a measure of socioeconomic status of studied families, but though no significant differences were found between high and low groups in ability to delay gratification, the authors do not rule out this explanation as they write that the measures employed were perhaps too “crude” (W. Mischel, 1958). These results between children with no father figure present and those with a father figure present were supported again by later research in the same cultural groups (W. Mischel, 1961), however, it was found that though these differences were statistically significant between groups at younger ages (8-9 years old) the differences did not reach statistical significance for older children (11-14 years old). Later research on children uncovered further relationships between ability or tendency to delay gratification with age and with intelligence (as measured by IQ tests) (W. Mischel & Metzner, 1962). The same research showed that the biggest change, from a preponderance of immediate gratification choices to a preponderance of delayed gratification for greater rewards, occurred between the ages of 8-9 (W. Mischel & Metzner, 1962).
Mischel also famously showed that the extent to which a child at 4 or 5 years old could delay gratification for a greater reward predicted that same child’s self-regulatory ability and subsequent academic and social success as an adolescent (W. Mischel, Shoda, & Peake, 1988; Shoda, Mischel, & Peake, 1990). These early indications at 4-5 years old continued to predict success 40 years later in finances and relationships (W. Mischel et al., 2010) and also predicted healthier Body Mass Index as adults (Schlam, Wilson, Shoda, Mischel, & Ayduk, 2013).
Self-regulation is proposed to be a facet of temperament (Rothbart et al., 2000) and therefore to be genetically inherited and a relatively stable trait. Temperament, from the biological makeup of an individual, dictates their biological sensitivities, their reactivity, excitability and impulsivity of physical and neural systems, with self-regulation being the neurological and physiological systems that modulate that underlying activity (Rothbart et al., 2000).
(King, Lengua, & Monahan, 2013) studied the contextual risk factors for interference in the development of self-regulation in the transition between childhood and adolescence, following 8-12 year olds for 3 years. They note that although the capacity for self-regulation was though to be developed only across childhood, recent evidence has shown that capacity for self-regulation continues to evolve at both physical and neural levels far into adolescence. It has also been shown that those who develop self-regulatory capacity more slowly during childhood and adolescence experience behavioural issues which involve lack of inhibition in externalising emotional issues.
A helpful definition of self-regulation is given by (Rothbart et al., 2000) as comprising two main facets: effortful control and impulsivity. (Rothbart et al., 2000) highlights two key aspects of effortful control; executive regulation of attention (for monitoring and perhaps also for strategic redeployment of attention) and inhibitory control, both used in order to supress a dominant response in order to enact a non-dominant, more correct or long-term adaptive response. Impulsivity though refers to approach behaviour, to speed of response, to surgency and exuberance or extraverted approach. So, effortful control employs cognitive, executive functions for self-regulation, while impulsivity employs approach, motivational drives for self-regulation.
(King et al., 2013) note that it has long been known that there is a genetically inherited component to temperament and hence also self-regulation, and that it differs between individuals, but that it has long been treated as a static capacity, an unmalleable trait that is perhaps revealed over time. However their research showed that it develops into childhood, and beyond it adolescence and that various factors can affect the speed of its development so that between individual differences can be observed.
Economic disadvantage has been repeatedly shown in research to negatively impact the progression of self-regulation development in children and pre-adolescents (King et al., 2013). Other home environment factors include parental practises. For example, a mother who is warm, sensitive and responsive, with positive expressivity and with firm structure to her relationship with her child will engender increased effortful control capacity in that child. This structure includes setting clear and consistent boundaries for the child as well as enforcing a kind of discipline that is not punishment or anger driven (King et al., 2013). In contrast, punitive discipline, power assertion, and coercion was associated in research with lower levels of effortful control in the child. The evidence for how much these effects linger beyond childhood and into adolescence has been lacking however.
In research tracking the differential progress of both effortful control and impulsivity development across childhood and into adolescence, (King et al., 2013) found that the two were uncorrelated, suggesting that they are separable constructs and may be influenced by different processes. It should be noted that harsh parenting was also associated with lowered impulsivity, which may seem like a positive outcome, but can reduce approach or motivated behaviour so that, rather than externalising behavioural issues, internalising issues emerge, such as depressive symptoms, anxiety and similar negative outcomes (King et al., 2013).
(Bernier, Carlson, & Whipple, 2010) built on research showing that early child-parent relationships (with research focus being especially on mothers) influence cognitive development. Their research showed that maternal sensitivity, mind-mindedness and autonomy support were positively associated with executive function development, including impulse control and other self-regulatory abilities. Parental sensitivity involves regular observance of and appropriate reactions to a child’s signals, which allows the child experiential success in affecting others in their environment. Parents should also allow children to solve problems, but with support structures in place, giving them age appropriate problems to independently tackle, but with support and by modelling problem solving strategies. Mind-mindedness refers to the parent’s use of mental vocabulary that is thought to enable children means of verbal expression and communication that make a smoother transition from external-regulation to self-regulation. Initial research shows that, among neural development, executive function development, including self-regulatory aspects, is especially dependent on the quality of appropriate input and sensitive interactions with a primary care-giver (Bernier et al., 2010; King et al., 2013). The research showed that these three positive parenting characteristics, including autonomy support, were positively associated with early infant development (3-6 months) and later (14-26 months) in working memory measures, conflict- and self-control measures. Autonomy support, where parents respect their child’s rhythm, scaffolding (by showing problem solving strategies) and allowing the child an active role in task completion, was the greatest predictor of executive function development (Bernier et al., 2010). While autonomy support is a large and stable predictor, mind-mindedness, after controlling for general cognitive ability, was the only predictor that contributed variance to executive function development at later ages (14-26 months), suggesting that the tools are used on the solid foundation that autonomy support builds.
Built on research that considers late infancy to preschool age, the first four years of life, to be integral to self-regulatory development, (Kochanska, Coy, & Murray, 2001) studied differences in this phase between genders in two arenas of self-control; committed compliance and situational compliance. Committed compliance, that is effortful control in genuinely trying to comply with a parent’s commands, was shown as being higher in girls than boys, who tended to exhibit more situational compliance, where compliance was given but with more reluctance or a lack of general commitment. Furthermore, the study showed that “Do” (enacting initiatory self-control) commands from parents were less successfully followed than “Don’t” (enacting inhibitory self-control) tasks. This is a pattern that seems to persist into adulthood. Committed compliance can involve an internalising of morals or reasons and be experienced more as self-regulation, rather than situational compliance which may feel more like external, pressured compliance that can result in issues later when self-regulation is needed (Kochanska et al., 2001). The research also suggested, albeit tentatively, that early committed compliance to a parent figure could generalise into better social self-regulation with other adults in emerging childhood.
(Posner & Rothbart, 2000) affirm that, especially given the wealth of recent research associating self-control with various life outcomes, the study of self-regulation (including self-control and broader systems) is crucial for understanding psychological development and also psychopathology. As (Posner & Rothbart, 2000) summarise, the neural substrates used in self-regulation are largely attentional networks and this should not be surprising, the maintenance, or achievement of thoughts or behaviours that align with set goals or standards can only be achieved by a constant monitoring and then adjustment of cognition or behaviour in relation to that standard. Attention can also provide priority to targeted tasks and this is seen in the increased activation of associated brain areas when increased focus is given to the related task (Posner & Rothbart, 2000).
Crucial landmarks that arise in the first year of self-regulatory development include more physical regulation such as orienting to objects and control of distress, whereas in the second year cognitive skills of regulation and planning emerge. In the first year of life, attentional networks, especially the anterior cingulate cortex and its monitoring of the experience of emotions, are used to control distress. The parent may externally regulate the infant’s distress by holding and rocking, and the infant begins to regulate its visual orientation, enabling the caregiver to try to distract the infant from distress by distracting its attention to some visual stimulus. This leads to development of the midfrontal brain region, essential for executive functions in general and especially for control of negative emotions. Early inhibitory control undergoes a massive increase in development in the third year of life (age 2-3 years old) as evidenced by the dramatically increased ability to inhibit a prepotent button press response in a modified Stroop task (Posner & Rothbart, 2000).
Temperament also shows emergence in the first couple of years of development, as those higher in effortful control exhibit more inhibition, sensitivity to punishment and withdrawal from excitement, whereas extraverted individuals show higher sensitivity to reward, and rapid approach behaviour to stimuli. These are thought to be basic, inherited biological features, however self-control can modulate them, by enacting approach behaviour in situations where the threat of some form of punishment is present, or enacting avoidance behaviour in situations where immediate reward is present. Early development in this area is highly predictive of later development, as individuals at 9 months who exhibit effortful control in restraint from touching a forbidden toy to a multi-task battery of effortful control assessment at 22 months (Posner & Rothbart, 2000). Various longitudinal studies using a similar battery of effortful control tests (one at 32-66months and one at 9-45 months) found that a child’s capacity for effortful control from 30 months and onward was highly consistent across tasks (suggesting that they measured a unified underlying construct general to all the self-control measures) and that this ability was highly stable across time (suggesting a trait-like quality, with levels of stability approaching highly stable traits such as intelligence or aggression) (Posner & Rothbart, 2000). Research by Mischel and associates extended the long-term stability and predictive ability of this trait beyond childhood, into adolescence and adulthood (Eigsti et al., 2006; W. Mischel et al., 1988; Shoda et al., 1990).
However, this is not to say that systems are set and unmalleable from early in the developmental process. Researchers used to believe that temperament had emerged and was set by very early in development and that it would change very little over time. Children’s reactive tendencies are observable early in life, but self-regulatory executive attention develops later in the first year and continues to develop through early school years (Posner & Rothbart, 2000). Various studies show that temperament is highly influential in personality and emotional development, but that self-control or effortful control is especially important in the development of morals or conscience, and in emotional, social and cognitive development (Posner & Rothbart, 2000). If temperament is foundational for personality development, then effortful control is important for the selection of elements as it focusses on what information to attend to and thus highlight in early years of development (Posner & Rothbart, 2000).
(Karreman, Van Tuijl, van Aken, & Deković, 2006) describe childhood self-regulation as the ability to manage levels of arousal and irritability, a construct that is greatly expanded in scope in later life. Self-regulation is traditionally seen as being based on temperament, which is itself regarded as being a biological, inherited trait, but self-regulation is also theorised to be shaped by an interplay of temperament with social experiences, especially with the chief source of social experience in the foundational first 2-3 years; the parenting practices of their mother, or primary caregiver. With the increasing number of studies examining parenting styles and self-regulation development, a meta-analysis was conducted by (Karreman et al., 2006).
Some early research found that controlling parental style was associated with improved self-regulation, while other early research found that controlling parental style was associated with self-regulation problems (Karreman et al., 2006). However, these seemingly contradictory findings can be explained by a finer distinction between positive and negative controlling parental style. Negative controlling parental behaviour is characterised by anger, harsh criticism, power-assertion, physical interventions and excessive or intrusive control. Positive controlling parental behaviour is characterised by guiding, with specific attempts at teaching, encouraging and directing rather than controlling. The directionality of this relationship is not clear however. It is possible that children with poor self-regulation cause their parents to act in a more assertive, controlling way, whereas those with good self-regulation allow their parents to be more helpful, encouraging and rewarding.
Because of some mixed results in the literature regarding the relationships between parenting style and self-regulation development (Karreman et al., 2006) conducted a meta-analysis of available literature. Positive control (defined as mild to moderate power-assertion along with clear guidance and encouragement) was positively associated with self-regulated behaviour. On the other hand, negative control (defined as stronger power-assertion, with coercive behaviour, critical comments and sometimes physical pressure) was negatively associated with self-regulated behaviour.
This effect of parenting in the development of elf-regulation is most starkly seen when children are not afforded secure attachments with a parent or primary caregiver (Kolk & Fisler, 1994). Research has shown that children who have been abused or neglected develop disorganised attachment patterns. This lack of stable, consistent, secure attachment with a parental caregiver seems to retard the normal development of self-regulatory capacity resulting in behaviours that may be maladaptive attempts at emotional and arousal regulation, such as aggression against others, self-destructive behaviour, eating disorders and substance abuse (Kolk & Fisler, 1994). Even when some form of early-childhood abuse or trauma is suffered, the presence of secure attachment with one primary caregiver can have powerful protective effects in mitigating damage and allowing development of self-regulatory strategies. Infants have limited regulation strategies to detach from overstimulation or over arousal initially (gaze aversion, self-sucking and dissociation) and rely on parents (often mothers) to give external regulation in the form of rocking, stroking, singing, feeding or verbalising. The ability for infants to detach from parents and explore stimuli and take in information, to increase stimulation and arousal, but to be afforded the ability to return attention to a secure attachment style of caregiving that can regulate their arousal and stimulation is important for the development of self-regulation, without the infant becoming overstimulated or under-stimulated. Because of responsive synchrony and mirroring behaviours the mother can also model self-regulatory strategies and begin showing the infant how their social interactions draw responses from external figures in their social environment.
Parents that are abusive or unresponsive to an infants signals of under or over stimulation can result in infants developing hyperarousal, which can result in disabilities in self-regulation of strong emotions. Primate studies with monkeys that have experienced parental abuse or neglect have shown that the following 3 or 4 years can result in aggressive behaviour, but with the therapeutic and normalising effects of being in a group of other normal peers, the effects can be largely reversed, although these reversals of poor self-regulation can come undone under high stress (Kolk & Fisler, 1994).
Parental issues with self-regulation or self-control can also affect a child’s development, via pathways such as parent alcoholism (Eiden, Edwards, & Leonard, 2007). Paternal or maternal alcoholism can affect the level of that parent’s warmth and sensitivity in responsiveness to the infant, which in turn can negatively affect the development of self-regulation. Simultaneously, this research (Eiden et al., 2007) also suggests that a child’s self-regulation capacity can mediate the extent to which a parent figure’s alcoholism engenders problematic externalising behaviour in children.
So, in Kopp (1982) model, between 12 and 18 months of age, children develop an emergent ability for control with an emerging awareness of social demands and the capacity to comply with parental commands. This in turn leads to a progressive internalisation of morals, as they become increasingly aware of societal norms and values. By 24 months children tend to have developed a basic self-control, allowing them to inhibit behaviour, despite tempting immediate rewards, or enact behaviour, despite imminent punishment, both in response to external commands but also in the absence of parental observation. By 36 months old the child exhibits a broader, more consistent system, called self-regulation, that can adjust to various situational demands.
(Kochanska, Philibert, & Barry, 2009) conducted a study looking at the interactions between genetic inheritance and environment (such as parental style) on the early development of self-regulation. They note that one area of focus is the serotonin gene regulatory region (5-HTT (serotonin transporter protein) LRP). Problems in the serotonergic system have been associated with regulatory issues in mood, attention and psychopathology. The serotonin transporter protein (5-HTT) has also been associated with neural substrates involved in executive function, including self-regulatory behaviour. Individuals born with a short variant of this gene have exhibited increased tendencies to depression, anxiety, and under-regulated, excessively compulsive behaviour (Eiden et al., 2007). In primates, this genetic risk resulted in high risk for various self-regulatory problems, but only if the infant primate had been separated from its mother. Human research seems to support this, with similar patterns in depression only being higher risk from the shorter genetic variant if the children had been exposed to neglect or abuse, and for anxiety or fearfulness, but only if children had received little social support.
This research (Eiden et al., 2007) confirmed earlier propositions that secure parental attachments can buffer against genetic susceptibility to maladaptive self-regulatory development.
(Kochanska, Murray, & Harlan, 2000) define that effortful control (elsewhere referred to as self-control) is separable though related to self-regulation, stating that effortful control is required in various processes such as self-regulation, emotional expression and socialisation. They also define that effortful control (or self-control) can involve both inhibitory control but also proactive initiatory or sustaining behaviour.
(Kochanska et al., 2000) review literature showing that attention and self-regulation in early development are strongly linked. Focussed attention that emerges at the end of the first year was proposed to be an antecedent for effortful control and this was supported by later research, which showed that greater sustained attention at 12-30 months predicted self-control at 24 months of age. This theory was supported in longitudinal research by (Kochanska et al., 2000). Further, research reviewed found gender differences in the early development of self-control, with girls scoring higher on assessments of social and behavioural inhibition, but with boys scoring higher on cognitive inhibition, findings which could be explained by reproductive and child rearing evolutionary adaptations versus hunting and providing adaptations. The study also found that parental personality traits such as self-regulation, prudence, patience, persistence and acceptance of cultural rules predicted infant effortful control. These parental traits are all clearly related to various aspects of self-regulation, yet the reason why these might predict child self-regulation are unclear. Genetic inheritance may play a role, as may the contribution of these characteristics to warm, sensitive, controlled parenting styles which engender self-regulation development, or it may simply be a case of modelling successful self-regulation behaviours. Though these findings were positive and statistically significant, their contribution to effortful control was modest, accounting for 11-13% of variation in infant effortful control. This suggests other contributory factors, with genetic inheritance undoubtedly playing a major role.
Other suggested factors (beyond genetics, secure attachment, parenting style, gender, early attentional capacity and parental personality) include other cognitive or neural characteristics (such as memory, language or speech ability and other attentional capacities) and other environmental antecedents (stressful events, stimuli in the physical environment, socialisations beyond parental influences, interparental conflict). Beyond these suggestions, the author suggests another potential antecedent that is, to the author’s current knowledge, not proposed anywhere in current literature and that is diet.
In adult populations, the differential metabolism of carbohydrates versus fats or proteins results in different curves and drops of energy availability. Carbohydrates, especially sugars and other refined carbohydrates, can result in spikes and drops of blood sugar which result in less even cognitive capacity. A mother’s diet during breastfeeding and a child’s diet during these crucial first 2-3 years could have a significant impact on the quality and consistency of energy and nutrients supplied to the brain, as well as the dynamics of energy and arousal swings in the body, potentially overwhelming emergent abilities for self-regulation if the spikes and drops of blood sugar are dramatic.
Some research has hinted at these relationships between early dietary practice and self-regulation but with other pathways suggested. (Birch & Fisher, 2000) report that the probability of becoming obese as an adult is over three times higher if one or more parent is obese. Research has shown that beyond genetic influences, there can also be phenotype influences on a child’s probability of obesity. Environmental effects on adiposity have been shown to be significant (Birch & Fisher, 2000) and within these influences there are shared or non-shared effects. Some research has suggested that non-shared environmental influences on obesity are significant whereas shared environmental influences may be negligible. Although it has traditionally been assumed that family environments must be shared between parents, children and other family members, it has been shown that there are in fact various non-shared elements that are experienced differently by different family members. One aspect of non-shared family environment influence on obesity is child-feeding practices between a mother and daughter and this was studied by (Birch & Fisher, 2000). The study found that although a significant portion of the variance in the high correlation between obese mothers having obese daughters was due to the biology of genetic inheritance, the model of best fit included additional variance explained by the mother’s child-feeding practices. Specifically these child-feeding practices were characterised by an overly restrictive feeding style, perhaps because of the mother’s own perceived lack of control and hence assumption that the daughter too cannot self-regulate adequately and also because of perception that the daughter too is at risk of becoming obese (Birch & Fisher, 2000).
(Graziano, Calkins, & Keane, 2010) studied longitudinal effects of various types of self-regulation at 2 years old and then again at 5.5 years old. They found that normative changes in BMI were predicted mainly by emotional regulation amongst regulation types, whereas paediatric obesity was predicted by emotional regulation as well as inhibitory control and reward sensitivity. BMI at 2 years old is a strong predictor for BMI at later years, but the above results were significant even after controlling for the variance explained by BMI at 2 years old.
(Kochanska & Knaack, 2003) studied various stages of child development from 22 – 73 months old. They found that effortful control measured at 45 months was highly stable longitudinally, becoming a traitlike quality. These children went on to develop behaviours suggestive of stronger consciences at 56 months and fewer externalising problems at 73 months. The longitudinal research also showed that by 3-4 years old this effortful control quality was as stable as IQ.
(Vaughn, Kopp, & Krakow, 1984) reported early results in the study of normative self-regulation development and individual differences, showing that early self-control development was associated with cognitive, language and socialisation development. However their results also suggested that with the tests used to measure effortful control, prior to 24 months of age, children were very inconsistent across time, task and situation and so concluded that individual differences did not emerge until after the second year.
Is there such a thing as “Food-Addiction”?
“Food-addiction” has long been mentioned in the general public and anecdotal evidence of it abounds, but it has only recently gained some support as potentially being an actual addiction in scientific research (Davis et al., 2011). However, one large problem and issue that distinguishes food addiction from drug addiction is that food is a necessary part of life, whereas drugs are not. Breathing oxygen could never be called an addiction, because it is simply a biological function that is required for life. The case could be made though that certain eating patterns go beyond normal requirements and thus may qualify for being addictions. The other potential point is that certain food items, or possibly entire food groups (carbohydrates) are not actually required for healthy human life (despite popular opinion).
This is reflected in a research article by (Tuomisto et al., 1999), the title of which mentions ‘Sweet Food “Addiction”’, with the quotation marks on “Addiction” hinting at the scientific disregard for such a construct at that time. In their study they measured various responses of “chocoholics” to see the extent to which they might correlate with behaviour patterns exhibited by drug addicts. Chocolate is a highly desired and palatable food, with many people self-identifying as chocoholics and anecdotally describing compulsive chocolate consumptory patterns that are reminiscent of drug behaviour. However, though previous research had found highly compulsive behaviour, on pharmacological grounds, the idea of addiction had been dismissed. However, other markers of addiction include changes in craving, affect and other physiological responses when presented with some cue of the target stimuli that is different from normal, non-addict populations (Tuomisto et al., 1999). In their experiment comparing reactions to self-identified “chocoholics” and control group individuals (Tuomisto et al., 1999) found that the reactions to first photos of chocolate, then a physical bowl of chocolates and then smelling, feeling and tasting the chocolates. Chocoholics salivated more than the control group, experienced higher desire to eat, and higher feelings of craving, excitement, frustration, anxiety, guilt and restlessness. Higher measurements of generally disordered eating and depression were also recorded in the chocoholic group. Chocoholics also consumed almost 4 times as much chocolate than the control group in a taste-testing task. Although experiential, self-reported differences were statistically significant, physiological differences, in the extent of salivation were different in only one of the two experiments, and heart rate did not differ between groups. Therefore the experiment offered some suggestive evidence for addict-like behaviour but was far from conclusive (Tuomisto et al., 1999)
An examination of the mixed evidence for and against the possibility of food addiction was presented by (Rogers & Smit, 2000). In the paper, the authors noted that there was little biological evidence to prove addiction, despite suggestions that addiction represented a specific nutrient need, or the presence of some naturally occurring psychoactive compound. As the author notes, defining the term “craving” is much easier (though not without its own problems) than defining “addiction”. Craving can be defined as an intense desire for something, or the feeling of a great or urgent need for something. But what can we define addiction as?
Drug “addiction” can be defined as being “restricted to the extreme or psychopathological state where control over drug use is lost” (Rogers & Smit, 2000). Another term that some say is more helpful and less confusing than “addiction, is “dependence”. “Dependence” describes the condition of “needing a drug or drugs to function within normal limits” (Rogers & Smit, 2000). Dependence is often accompanied by symptoms of tolerance, sensitisation, withdrawal and craving. Tolerance describes the decreased sensitivity to the effects of a drug, requiring the user to take larger and larger amounts to get the same effect. In other cases though, repeated use can lead to the reverse effect, a heightened sensitisation to the drug. Withdrawal symptoms occur when the drug use ceases for some time and usually involves symptoms that are exaggerated opposites to the effect of the drug. For example, withdrawal from heroin can involve dysphoria (rather than euphoria), agitation (rather than relaxation), nausea and cramping (Rogers & Smit, 2000). One feature of addiction that some argue is central is that it makes neural changes which affect behaviour, cravings and sensitivity to rewards. Some expand the definition to include any behaviour (consumption or otherwise) that one is “given over to”, which involves consistent or chronic compulsive action despite repeated, serious resolutions to stop. Another key feature is that this compulsive behaviour is significantly damaging to the individual’s life and that despite this harm to self is resistant to attempts to cease. The individual chronically pursues reward and/or relief by substance consumption (or by certain behaviour).
Depending on which parts of an addiction definition one chooses to accept or reject, food such as chocolate may align completely or only partially with being a potentially addictive substance. One area that has caused doubt in labelling chocoholics as addicted is the seeming lack of any psychoactive substances in chocolate, or ability to create any psychoactive effects or neuroadaptive changes in the user (Rogers & Smit, 2000).
Various articles have reported that consumption of chocolate leads to improved mood, although there are mixed reports of ambivalent mood changes or even negative shifts and guilt or depressive symptomatology (Rogers & Smit, 2000). In any case, mood changes alone are not evidence of psychoactive qualities, as any food, or any activity can cause mood shifts. Reports, especially in popular media, have abounded of cocoa-based “brain cannabinoids” in chocolate, such as relatively high concentrations of theobromine. However, this is a relatively mild stimulant and does not produce strong subjective effects on the central nervous system (Rogers & Smit, 2000). Another compound touted in chocolate is methylxanthine (caffeine), however the concentrations found in chocolate are significantly weaker than in tea or coffee. However, there has been no association observed between compulsion for chocolate and compulsion for tea or coffee. Other compounds suggested as psychoactive in chocolate include tyramine, phenylethylamine, serotonin, tryptophan, and magnesium, however all of these substances are found in significantly higher quantities in other foods, which do not produce the same appeal as chocolate does.
A cannabinoid (substance that acts on cannabinoid receptors in the brain to produce an effect, often high-like) called anandamide was discovered in chocolate and other cocoa based products, however its concentrations are so small, that a 70kg man would have to consume 25kg of chocolate in order to achieve any noticeable high-like effects (Rogers & Smit, 2000). A further note of caution, as described by (Rogers & Smit, 2000), is that the discovery of a psychoactive compound in a food does not necessarily lead to the conclusion that it acts on the brain to produce psychoactive effects.
Caffeine has been one of the most often presented evidences of food addiction, especially with coffee and to a lesser extent, tea. At least a mild dependence on coffee or other caffeinated beverages has been discovered, with withdrawal symptoms such as headache, fatigue and sometimes even mild nausea (Rogers & Smit, 2000. Interestingly though, one marker of addiction “craving”, is usually absent or only very mildly experienced by regular caffeine users, who rather enjoy it is part of a habit or routine, rather than a strong craving driven compulsion. Some research has shown that even when caffeine use is reduced due to situational or schedule interruptions, the user often does not register any “need” or “craving” for coffee {Rogers, 2000 #1831).
One study provided chocolate cravers with various versions of chocolate, white, milk and coca capsules. The actual chocolate bars satisfied cravings, whereas the coca capsules (which would presumably have contained the potential psychoactive compounds in greater concentration) did not, suggesting that it is not primarily any psychoactive, or drug like properties of chocolate, but rather a combination of smell, touch, memories, experience, social and psychological associations, taste, smell and mouth-feel that elicit the effects of chocolate consumption. This is echoed in statistics that show that the most highly consumed form of chocolate is milk chocolate, or milk chocolate covered/containing confectionery. Milk chocolate is also regularly reported as the highest rated by self-identified chocoholics. However, milk chocolate contains significantly lower levels of cocoa than dark chocolate and thus less potentially psychoactive compounds. This suggests two things; that it is more about other elements of the overall consumption experience as described above, and that the sugar or fat content may be more responsible for cravings than any cocoa based substances.
Carbohydrates in general, and perhaps most powerfully refined sugar, have been observed in research as being used to elevate perceived flagging in energy levels and also to elevate depressed mood (Chao, Grilo, & Sinha, 2016; Wurtman & Wurtman, 1989, 1992, 1995). Coffee has traditionally been used to sharpen attention or elevate energy levels. It has been theorised that meals high in protein increase alertness, whereas those high in carbohydrates elevate mood, especially in “carbohydrate-craving obesity”, premenstrual syndrome and seasonal affective disorder, where the carbohydrate may be consumed in order to affect brain serotonin neurotransmission. As impaired serotonin function has been implied in the aetiology of depression, this consumption of carbohydrates can be viewed as a kind of self-medication, whether intentional or not. For this reason, “carbohydrate-craving obesity” may be viewed as a kind of mood disorder, where appetite control and physical health are sacrificed to medicate the mood dysregulation (Rogers & Smit, 2000). However, though these ideas are in the popular scientific media, and have some suggestions in scientific literature, the case is not sufficient and, like chocolate, the observed fluctuations in serotonin are too subtle to theoretically have a large impact.
Cravings for food that is high in sweet carbohydrate and/or fat has been linked with the opioid reward system, rather than the serotonergic. Therefore if there is an addictive quality to some food consumption it may share neural pathways with opiate drug addiction. The release of these opioids may be, like the proposal regarding serotonin, a kind of self-medication, in this case to relieve stress. This idea has been supported by research which shows that opioids are released in response to the ingestion of sweet foods, that opioid antagonist drugs reduce food intake in binge eaters by reducing the consumption of sweet foods such as chocolate and biscuits and that opioid antagonist medication may reduce stress-induced overeating (Rogers & Smit, 2000).
Food intake changes during the menstrual cycle have been observed with increased reports of cravings during the premenstrual phase of the cycle and during pregnancy. The premenstrual phase of the cycle also sees a peak in caloric intake, whereas the follicular (preovulatory) phase sees the biggest drop in caloric intake. Possible explanations include changing nutrient needs during the cycle, as well as attempts to attenuate mood changes brought on by hormonal shifts in the cycle.
Bulimia (marked by episodes of bingeing and then purging) or general binge eating can share many symptoms of drug addiction. Episodes of bingeing involve a compulsive consumption of large amounts of food (typically over 3,000kcal) usually composed of high carbohydrate and high fat food or drink. The individual experiences a lack of control over their behaviour, sometimes even to the extent of something akin to a fugue state, after which the individual will “come around” or snap out of it and wonder what they have just done. As with substance abuse addictions, this behaviour can have a large impact on the individual’s life and wellbeing, with attempts to hide their behaviour, extreme reactions of guilt, shame (and purging) after episodes, depressed or anxious mood, pathological loss of control, avoidance of social situations that interfere with the target behaviour and a restructuring of a daily schedule to enable the next “fix”, to the extent that an individual’s life revolves around the rhythm of the addictive behaviour.
When discussing the extremity of binge-eating and the resultant purging (which may come in the form of induced vomiting, fasting, extreme dieting, or extreme exercising) it is based on the assumption of modern society that 3 square meals a day, providing a steady, moderate caloric intake is the normal pattern that is most adaptive for humans. However, for the bulk of homo sapiens sapiens history, food intake was likely much more uneven, something more akin (though perhaps not as extreme) as a binge-purge pattern, with times of feasting after a bountiful hunt and then fasting until the next. This may have been on a daily basis (mornings and mid-days spent subsisting on no or meagre foods from foraging, until the evening where hunted protein and fat could be consumed) to longer days of foraged food subsistence during leaner times. This indicates that appetite control system may not be as rigidly fixed as previously theorised and is in fact flexible in allowing for these uneven times of caloric opportunity. Therefore the homeostasis of appetite may not be as tightly controlled as previously thought.
Depletion and repletion cycles (measured by the body in terms of physical fullness, nutrient content and others) may contribute to appetite and consumption motivation but so too may the existence of energy resources kept on the body in the form of body fat. It has been noted that the effect of feedback from the level of body fat (energy reserves held) is quite weak (Rogers & Smit, 2000), however this observation is perhaps weakened by the underappreciated fact that energy reserves and accessibility of energy reserves are separable. Many people that store large amounts of body fat, because of the effect of insulin, paradoxically have their stores shut off, or very difficult to access for metabolism into energy and in fact preferentially store more of it. Therefore, short of them entering a ketotic state wherein their body fat reserves would be preferentially burned, the bioavailability of their body fat is minimal and thus it having a minimal effect on appetite should not be surprising.
Although physical energy levels influence appetite, research has showed that appetitive cues can fade and have less influence later in the day if initially resisted, again reflecting the body’s appetite control flexibility, and that external cues can have a powerful effect on appetite. Although the amplification, or muting, of external appetitive cues is influenced by the individual’s caloric and nutritional depletion/repletion states, external stimuli can prompt consumption regardless of physical need. External cues can provoke increased salivation, insulin release and gastric acid secretion which are then fed back to the brain and interpreted as hunger and a desire to eat that specific food.
During consumption, experiences of hunger and desire to continue eating are constantly controlled by stimulatory and inhibitory inputs. The feeling of food mass entering the stomach and triggering stretch receptors, as well as chemical receptors picking up the intake of various nutrients, have an inhibitory effect on experienced hunger. The oro-sensory stimuli of the food (the mouth feel of the food) as well as visual and scent cues have a stimulatory effect to continue eating. The palatability of the food contributes stimulation to continue eating and is a combination of innate taste preferences (for example for sweet food) and learnt tastes.
The physical input of information is weighed constantly to give an overall, a net drive to consume or not. Yet at times, in the face of appetitive urges to consume, dietary restraint may be exerted. Individuals probably contain both biophysical and cognitive “set-points” regarding their body weight, size and shape. A biophysical set-point is probably constructed from an inherited foundation of what is “natural” (based on genetic variables such as stature, bone structure, gender and so on) in combination with hormonal and metabolic influences which may fluctuate over a lifetime. Learnt behaviour, cultural norms and other environmental influences may also shape an individual cognitive set-point for an individual’s “ideal” body. The individual’s self-regulatory systems monitor current state and alert the individual to any deviations from the set-point (whether biophysical or cognitive). When a deviation from the desired set-point is observed, regulatory behaviour will be motivated and perhaps initiated.
What exactly constitutes a normal set-point, a normal desire for food and normal eating patterns is difficult to define. Therefore, extreme deviations from this that could be labelled “cravings” “food addiction” or “dependence” are thus also difficult to define. When does simply wanting to eat a bowl of cereal constitute an addictive-like craving? It is a normal, culturally accepted food to eat at breakfast in most Western countries. But does wanting to have it at night constitute an addictive “craving” because it is “abnormal” or beyond the culturally normative boundaries? Does wanting to have a bowl after a filling dinner constitute a craving because it is beyond normative ideas of “enough food”? Therefore, because food in general is a necessity in life, whether the particular eating pattern is viewed as subject to problematic “cravings” (versus adaptive hunger), or “addiction” (versus natural, necessary eating to survive) can be heavily influenced by the cultural, learnt lens through which it is viewed. For example, both processed breakfast cereals and chocolate can be highly caloric, high sugar foods with little other nutritional value and yet chocolate is often seen as the object of compulsion and cravings, whereas cereal is seen simply as something to eat for breakfast when you are hungry.
The restraint often exercised for certain foods, such as chocolate, can lead to enhancing the desirability of that food as a “naughty” or “taboo” food, such that the simple act of eating some calories can become entangled with various thrills, rewards, guilt, shame, excitation and later depressed mood. The association of these ambivalent emotions tied to certain foods can lead to cycles of indulgence and resistance that can contribute to the aetiology of bulimic eating disorders (Rogers & Smit, 2000). As these associations between certain foods and cultural taboos implies, the terms addiction and craving, although certainly built on biological foundations and neural changes, cannot be understood as maladaptive unless seen that way in a cultural, learnt context, just as with drug addiction.
To the extent that drugs are readily available (especially as is the case with smoking and drinking alcohol), some researchers (Rogers & Smit, 2000) view substance addiction as an automatic process, built on strong neural pathways of cognitive and motor behaviour that is often carried out almost without the individual noticing and that urges, cravings or any other emotional reaction only occur when an attempt is made to resist the automated behavioural patterns. Some people report similar patterns with food; they simply eat, mindlessly and only experience strong cravings or urges when they are actively trying to resist their habitual use, or in a situation that otherwise prevents them from getting their “fix”. With both food and drugs, the times of failed resistance and subsequent shame, or anxiety when confronted with a temptation, are indicative of an ambivalence toward the target of addiction, or addiction-like feelings. The individual often both loves and hates the food or drug, and derives from it both intense reward and intensely negative consequences.
One potential explanation for the prevalence of food-addiction in the public consciousness is the very prevalence of food-addiction in the public consciousness. Attribution refers to the attempt of an individual to apply some concept to a behaviour which they do not understand, in an attempt to understand it. Because food-addiction is such a well-known concept, as are cravings, when an individual experiences compulsive, automatic behaviour that they cannot explain, and that is counter to their schema of what their eating should be like, an attribution of food-addiction may be made for its explanatory power. This neither proves or negates the possibility that food-addiction is a scientifically viable construct, but merely suggests a means by which it may be experienced by individuals.
With this in mind, it is advisable to reconsider taking people’s self-reports of cravings and chocolate or other food) addiction, at face value. An individual’s self-perception of addiction can be strongly heightened by the level of taboo on that food, by their learnt emotions around it, and by their attempts to restrain consumption, and thus the ambivalent attitudes built around it. One individual may simply eat one or two bars of chocolate a day and feel no craving, compulsion, ambivalence or guilt about it. Another individual may feel that chocolate is taboo, should be eaten only in very small quantities or not-at-all, and be actively engaged in restraining their consumption, thus giving them an experience of lack-of-control, ambivalence and overconsumption. Therefore, any self-reports of “cravings” (With the implication of being illicit) or food “addiction” should be viewed in their context of individual relativity and learnt, cultural context and attendant attitudes.
In her review of the literature(Marcia L Pelchat, 2002) defined a food craving as an intense desire to eat a specific food. As she noted, this definition carries two important distinctions; one, that it is intense and thus beyond regular food choices, and two, that it is for a specific food and thus is differentiated from regular hunger. Food cravings are extremely common, as almost 100% of young females and 70% of young males sampled had experienced food cravings in the past year. However, in older samples the rate of food craving appears to decrease, a result which mirrors that found in drug addiction. Research showed that when fed a nutritional shake as the only source of nutrition, participants experienced greatly increased cravings, especially for foods that differed in sensory and nutritional make-up ways from the shake being consumed. This suggests that one function cravings may serve sometimes is to induce the organism to a greater nutritional variety.
Just as the pattern of lessening food cravings over lifespan mirrors that seen in drug cravings in addicts, so too do other elements of food consumption ad drug consumption. A large body of research literature has established an association in animals (usually rodents) between preference for sweets and self-administration of drugs (Marcia L Pelchat, 2002). Addiction programs in humans have also traditionally recommended the use of sweets to buffer the cravings for drugs and this has been reflected in research, showing that opiate drug and alcohol recovery patients exhibit a heightened craving for and consumption of sweet foods, when compared with non-prior-substance abuse samples (Marcia L Pelchat, 2002). Further, although there is little evidence for inheritability of preference for sweet food in humans, some research suggests that children of alcoholics express an increased preference for sweetness.
These correlates may be explained by overlaps in neural pathways. Although the endogenous opiate system is responsible for addiction to opioid drugs such as heroin, methadone, morphine, codeine and others it is also implicated in the reward effects of alcohol. Alcoholics who have been regularly given opioid receptor antagonists exhibit improved rates of abstinence, reduced cravings and reduced overall alcohol consumption. Just as opioid drugs are often used as painkillers, there is some evidence that sweet food consumption may reduce pain and that this analgesic effect may be due to the production of opioids neutrally. When babies were given sugar water to drink before a blood test prick in their heels they cried less (Marcia L Pelchat, 2002). This has also been proposed in breast milk, which is predominantly sweet and has analgesic properties. Much of the research suggesting analgesic and opioid effects of food has been conducted using sweet food, so it is unclear whether this is a property solely of the sugar in those foods or if any palatable food would give similar results. There has also been some research suggesting that the administration of opioid blockers to binge eaters can reduce their cravings in general, whether for sweet food or other (Marcia L Pelchat, 2002).
Serotonin too has been implicated in drug addiction, as brain serotonin (5-HT) levels seem to interact with intake of alcohol, morphine and cocaine (Marcia L Pelchat, 2002). Some research has shown that ingestion of carbohydrate (in the absence of protein) increases the bioavailability of tryptophan, the precursor to serotonin, which may lead to an increase in brain serotonin. This may explain the observed tendency for those with depressive symptoms or those with depression to crave carbohydrates as a kind of serotonin increasing self-medication {Pelchat, 2002 #1838}. Further evidence for this has been shown in patients administered with serotonin raising drugs who subsequently experienced lowered intake of carbohydrates (Marcia L Pelchat, 2002).
Dopamine is responsible for the rewarding nature and salience creation for alcohol, drug and various other behaviours, but the research literature on the effects of dopamine drugs on food addiction or cravings is sparse.
The overlaps between disordered behaviour, neurotransmitters and addiction-like food behaviour are also seen in serotonergic drugs which improve both obsessive-compulsive disorder and eating disorders. The orbitofrontal cortex, which is implicated in obsessive-compulsive disorder, is also associated with cocaine and alcohol cravings. This area of the brain, which also receives neural pathways from reward centres, might also be responsible for the obsessive thoughts for food and compulsive consumptory behaviour of those who self-identify as addicts, but there is no direct evidence for this currently. However, high comorbidity between eating disorders and obsessive-compulsive disorder suggest this brain region may also affect dietary behaviour, as does the higher than normal carbohydrate craving in those with obsessive-compulsive disorder (Marcia L Pelchat, 2002).
Research indicates that, just as in drug addiction, food craving does not have to arise from an imbalance to homeostasis, or any kind of nutritional balance, nor does the nutritional or pharmacological intake of the craved food need be required to satiate, rather the taste alone can achieve this craving satiation. Also, the triggering of food cravings, as with drugs, is not dependent on internal metabolic states or hunger states, but is often triggered by external cues, or by stress or emotional cues. These lines of evidence indicate there is a significant learnt element, of paired associations between the craved food and some reward, an incentive salience created via dopaminergic pathways (Marcia L Pelchat, 2002).
A commonly cited element of addiction-like food cravings is sugar, or sweet foods. (Avena, Long, & Hoebel, 2005) used rats to show that animals may be susceptible to neural changes and dependence on sugar consumption. They offer a helpful definition of addiction as a pattern of compulsive, uncontrollable behaviours that interfere with normal activities and note that the diagnostic criteria for addiction involves three stages. The first is bingeing, which marks a transition from substance use to substance abuse, with escalating intake and unusually high intake at one time. This leads to the second stage, which appears when the binge is resisted or is unable to be achieved by situational demands resulting in withdrawal. Withdrawal symptoms are dependent on the substance of abuse but are often mirror images of the effects of the abused substance, such as anxiety, depressed mood, somatic pain or headaches. The third stage, craving, similar to the second, becomes apparent after a period of abstinence. It is a heightened motivation and drive to obtain and consume the substance in the face of deprivation. It often occurs after a period of intentional or forced abstinence or attempts at restraint. One measure of craving intensity is enhanced effort to obtain the substance after a period of deprivation (Avena et al., 2005).
(Avena et al., 2005) note the behavioural and neural overlaps and comorbidities between drug addiction and food intake, especially of sweet food. With experiments on rats, a group that was allowed heightened consumption of sugar were then subjected to 2 weeks of abstinence. Upon renewed availability of sugar, the rats exhibited significantly increased consumption, indicative of craving and dependence, when compared with a control group. (Rada, Avena, & Hoebel, 2005) note that rats allowed daily access to sugar or regular chow exhibit neural and behavioural changes similar to drug dependent rats. When deprived of sugar for 24 hours they show symptoms of withdrawal and when reintroduced to sugar show signs of craving and dependence (Rada et al., 2005). With similar neurochemical markers to most drugs of abuse, rats show increased dopamine in the nucleus accumbens with sucrose bingeing and show neurochemical adaptations such as sensitisation, withdrawal and tolerance along with a heightened dopamine response to sucrose after repeated bingeing (Rada et al., 2005).
As (Avena, 2007) notes, binge-eating may be maladaptive in modern times when food is ubiquitous, but may be an evolutionary adaptive skill from times when food source availability was more uneven and unreliable. Not only has the constant supply of food meant that what may be an innate formerly adaptive trait has now become problematic, but the nutritional content of food has also changed, meaning a level of sweetness and fat in many foods that are relatively devoid of any other nutritional value as evolutionary sources of fat or sweetness would not have been. Binge-eating is not confined only to clinical eating disorder populations, as at least 6.6% of the non-clinical population have been suggested as bingeing (Avena, 2007), a behaviour that can lead to obesity. In animal models, as described above, (Avena, 2007) had established eating behaviours and neurochemical changes similar to drug addiction in rats, for sugar.
In human samples, (Cassin & von Ranson, 2007) studied the similarities and differences in experiences of addiction-like feelings in a binge-eating disorder sample. In a sample of 79 binge-eating disorder women, most (92.4%) met the DSM-IV criteria for substance dependence, but when a second, stricter, diagnostic criterion (Goodman’s) was applied, 40.5% of the women qualified for a diagnosis of addictive disorder (Cassin & von Ranson, 2007). They show that in clinical settings, 26.9% of clinicians sampled treat eating disorders with addiction-based therapies. They also note that the addiction model of eating disorders classifies binge-eating as a physiological food addiction, and that animal models show shared neural pathways and changes based on sugar dependence. They also note the difficulty in defining food bingeing as an addiction because there is no consensus on what defines an addiction, and in fact, no diagnostic manual currently utilises the construct “addiction” (Cassin & von Ranson, 2007). In their application of the DSM-IV and Goodman’s criteria they discovered some addiction-like traits within the studied sample. For compulsive behaviour, most women reported consuming much larger amounts of food than intended, continuing to engage in consumption despite knowledge that it had adverse effects on their health and life, and having a persistent desire, but with many unsuccessful attempts, to cease the behaviour. Symptoms of tolerance and withdrawal had more variability between participants, with common withdrawal symptoms including irritability, moodiness, anxiety, restlessness, migraines, insomnia, poor concentration, and lethargy. The author’s conclude however by suggesting that, despite many areas of overlap, those with binge-eating disorder also have body shape and weight concerns, which makes their behaviour more intentional in some ways and makes the classification as addiction more problematic. Because of this, and the fact that depending on whether the DSM-IV or Goodman’s diagnostic criteria for dependence/addiction are used, between 8-60% of binge-eaters may not be classified as having a food addiction. They therefore recommend that it remain classified as an eating disorder rather than a food addiction.
(Avena, Rada, & Hoebel, 2008) present a logic based argument in their introduction before presenting their research on animal models of food addiction. They state that the same neural systems (opioid and dopaminergic) that evolved to motivate seeking food and then reinforce food seeking behaviours are the same systems that underlie drug addiction. Therefore, (Avena et al., 2008) states, it seems logical that food could also be prone to these neural systems of addiction. Their animal (rat) based research has shown mild, but well-defined dopaminergic effects of intermittent access to and resultant binges of sugar. When sugar is not accessible, or when naloxone (an opioid antagonist) is administered, then withdrawal symptoms are observed (Avena et al., 2008). In the rat subjects, all markers of addiction were observable after being allowed intermittent access to sugar along with regular chow. Bingeing behaviour, withdrawal symptoms and then craving (marked in rats by intense acquirement behaviour) were created in the sugar group. Sensitization was also observed when, after a period of abstinence a small amount of sugar was allowed. For the control group, this small amount of sugar had a minimal effect, but for the sugar dependent group, marked hyperactivity was observed. Along with the well-established dopaminergic and endogenous opioid system involvement in sugar dependence in animal models, the effects on another neurotransmitter, acetylcholine, has also been observed. Increases in acetylcholine, while dopamine is low (due to the lack of dopamine enhancing drug or food rewards) contributes to withdrawal systems including somatic, muscular symptoms and depressive symptomatology.
Based on their animal models (Avena et al., 2008) conclude that in humans, sugar can become a source of addiction when consumed in a “binge-like” way. They note however that the intensity of the addiction symptoms to sugar (bingeing, withdrawal, craving and sensitisation) is lower than that observed in drugs such as cocaine and that as yet they have not tested the extent to which rats might endure physical obstacles or pain to acquire sugar fixes (as they have been shown to do with cocaine). Therefore it is unclear whether humans might exhibit behaviour that could warrant the neglect of social activities in service of sugar consumption as described in the DSM-IV (Avena et al., 2008). In bulimics, the pattern of binge-eating is typically a fast, or minimal caloric intake during the morning and day with large binges generally occurring at night. Bulimics have also been shown to have low central dopamine activity and dopaminergic and opioid responses similar to that of drug addicts. Some studies have also experimented with fat bingeing in rats, and though some bingeing affects can be induced, the results are more mixed than with sugar, with little or no signs of withdrawal or dependence (Avena et al., 2008).
(Marcia Levin Pelchat, 2009) reports that the distinction between wanting and liking is important to remember in addiction, as many drug addicts continue to crave and want the drug, but have long since ceased to enjoy or like it. In food however, wanting and liking are generally seen to overlap, but there are some cases where they may be distinct for certain foods with certain people (Marcia Levin Pelchat, 2009). (Marcia Levin Pelchat, 2009) cites recent research showing that just as cocaine causes release of dopamine, the same is true for food. Opioids also support the rewarding quality of food consumption, drug consumption or other behaviours but also contribute feelings of pleasure and is also increased in food, especially sweet and palatable food, consumption, though it does not appear to impact hunger or satiety signals (Marcia Levin Pelchat, 2009). In an fMRI study, (Marcia Levin Pelchat, 2009) used a subtractive process (subtracting imagined or thought about foods and liked only foods) to selectively highlight brain areas activated by food cravings in participants. The hippocampus, insula and caudate were all highlighted, just as seen in drug cravings, which lends initial support to a theory of a shared neural substrate for food and drug cravings (Marcia Levin Pelchat, 2009). Similarly to external stimuli and learnt associations triggering drug relapse regardless of withdrawal symptoms, food cravings can also be triggered by external stimuli that have been learnt as associated with the target food, regardless of hunger or motivation states.
As (Gearhardt, Corbin, & Brownell, 2009a) states, animal models, clinical observations and general use of terms promote the idea of food addiction being a real construct, however the scientific evidence is equivocal in humans.
The diagnostic criteria for Substance Dependence (not called addiction) in the DSM-IV-TR, and requiring that 3 of the following 7 criteria are met and that they result in clinically significant impairment or distress, were as follows;
- Tolerance, as defined by either of the following:
The need for markedly increased amounts of the substance to achieve intoxication or desired effect.
Markedly diminished effect with continued use of the same amount of the substance.
- Withdrawal, as manifested by either of the following:
The characteristic withdrawal syndrome for the substance.
The same (or closely related) substance is taken to relieve or avoid withdrawal symptoms.
- Taking the substance often in larger amounts or over a longer period than was intended.
- There is a persistent desire or unsuccessful effort to cut down or control substance use.
- Spending a great deal of time in activities necessary to obtain or use the substance or to recover from its effects.
- Giving up social, occupational, or recreational activities because of substance abuse.
- Continuing the substance abuse with the knowledge that it is causing or exacerbating a persistent or recurrent physical or psychological problem.
The first criteria, tolerance, though not necessary, depending on the number of other criteria present, has received little research in food in humans. Withdrawal for sugar has been observed in rats but not yet in humans. Tolerance too is difficult to study in human models because of methodological issues and because of the generally early initial intakes of target foods for substance abuse, such as sugar. However, there is some preliminary suggestive evidence for tolerance effects in some foods or in some situations. For example, in binge-eating bulimics, as the duration of the eating disorder continues the frequency of binges and the amount of food eaten tends to increase (Gearhardt et al., 2009a). A further suggestive result of tolerance is the building insulin tolerance, or insulin sensitivity blunting effect of prolonged sugar consumption, seen in the development of metabolic syndromes or Type-II diabetes.
As discussed previously, withdrawal symptoms from sugar dependent rats has been well established, however there are only suggestive results in humans. Examples of this are blood sugar drops, reactive hypoglycaemia, occur as reactions to sudden and dramatic drops in food (especially sugar) intake or restriction. Symptoms of discomfort, lightheadedness, great hunger, fainting occur. These symptoms are common in those not in a ketotic state, that is, the majority of individuals eating a typical modern diet that is carbohydrate dependent.
Taking the abused substance for longer than intended or in greater quantities than intended indicates a loss of control that is often seen in binge eating episodes (of up to 5,000 calories an episode) of bulimia nervosa (BN) or binge-eating disorder (BED). Although these two clinical diagnoses only apply to 1-4% of the population, a further 9% of the non-clinical normal weight female population and up to 21% of the non-clinical overweight female population have reported occasionally binge eating (Gearhardt et al., 2009a). Beyond binge-eating episodes that are easily recognisable as moments of lost control, consistently overeating, either in terms of calories or in types of food, is becoming increasingly normative, as evidenced in the continued weight gain of modern populations. There is evidence that some carbohydrates, especially refined and sugar (and especially fructose) did not trigger satiety signals as effectively as protein or fat do and thus may lead to a consistent, systemic dysregulation of appetite and healthy food consumption. The non-episodic nature of this last point though make it difficult to delineate as a clearly marked period of lost control, as is possible in drug or alcohol abuse.
Criteria 4, the repeated desire or effort to cut down or cease consumption of the substance of abuse, is perhaps the most prevalent and easily observed in food consumption of all the criteria. Individuals in the US report that attempts to diet for health or weight loss are the most common goal, but that loss of control, lack of willpower and failed attempts are the norm (American Psychological Association, 2012). It was estimated that up to 33 billion dollars was spent annually (in the US? What yearS?) on dieting products and programs, and that this desire to cut down or restrict some food intake begins early, with 37% of elementary school children in a US sample reporting dieting behaviour (Gearhardt et al., 2009a). The continued increase in weight, the continued use of diet programs, and empirical data reporting that 83% of dieters regain the weight and more within 5 years (Gearhardt et al., 2009a), all indicate that the majority of attempts to control intake of some quantity or type of food is repeatedly unsuccessful and is also consuming considerable time, energy and financial resources of a large portion of the population throughout their lifespans.
This great use of time, energy and financial resources spent on the attempts at food control address criteria 5, the great deal of time spent in activities to use, obtain or recover from the substance of abuse. One criticism is that because the most common target for food addiction, sugar, is so cheap and readily available and requires minimal preparation in fast-food forms, that the time spent to use and obtain food discount it from being an addiction. The note above indicates that the time spent on recovering from the substance of abuse is prohibitive and that in fact the legality or ready availability of nicotine is similar to that of food. There are anecdotal reports of individuals going to extended lengths to have a certain type of food that they are craving, even though other sources of nutrition are more readily available, whether in terms of cost, distance or social expedience. Binge-eating disorder (BED) and bulimia nervosa (BN) involve behavioural measures that avoid certain social situations in order to serve their cravings and indulge in binge behaviour.
These activities, commonly seen in BED and BN address criteria 6, giving up other social, occupational or recreational activities because of substance abuse. The results of maladaptive food consumption are clear in WHO estimates of work days lost from various mental and physical disorders that have poor diet as a risk factor. Body shame and self-consciousness due to being overweight are common reasons for people to avoid social activities. Further, the lost energy and fitness often result in reduced physical and social activities. Laboratory tests have shown that obese individuals will go to greater lengths than normal weight individuals to obtain sugary snack foods, rather than engaging in a pleasurable sedentary activity (Gearhardt et al., 2009a).
These issues also address criteria 7, which requires continued use despite physical or psychological problems. The knowledge of dietary effects on weight and health conditions is increasingly available and yet weight and health issues continue to increase. The health costs are compounded by financial costs in doctor’s visits and treatments for resultant issues. Further to these costs are the social stigma suffered for being overweight or for indulging in binge eating, whether at a clinical level or not. Trials on diabetes treatment and a recent one on heart health (from which 139 participants were dropped because they could not abstain from chocolate for the trial period) further highlight this problem of control for some individuals despite clear costs (Gearhardt et al., 2009a).
Despite firm evidence for some criteria and suggestive evidence for others, the further caveat in the DSM-IV-TR diagnostic guide is that these must involve “clinically significant impairment or distress”. What reaches the level of clinical significance is often a subjective judgement and difficult to clearly delineate. The level of impairment is highly individualised as is the level of distress caused. However, one study of 4283 overweight individuals by Schwartz (from (Gearhardt et al., 2009a)) highlighted the distress felt with their inability to control their diet and weight. 46% reported that they would give up a year of their life instead of being obese, 15% would give up 10 years of their life, 30% would rather be divorced than obese, 25% would rather not be able to have children than be obese, and 14% would rather be alcoholic than obese. Furthermore, some of the figures previously discussed, about years of life lost, working days lost and amount of money or time spent on trying to control or quit certain food intake suggest a significant level of distress.
Although it could be argued that the amount of impairment or distress caused by alcohol or drug intoxication can be more dramatic and starkly contrasted against normative behaviour in society, this in fact highlights one danger of dismissing food addiction. The contrast between drug or alcohol abuse and normative societal behaviour is stark partly because it is the exception to the rule. However, health, social or occupational distress or impairment, though perhaps lower level in many cases and accepted as normal, are perhaps more worrying precisely because they have become normative to some extent.
(Ifland et al., 2009) propose an addiction model based on refined foods, notably sugar, flours, fats, salt and/or caffeine, and note that as these food elements have increased between 1970 and 1997 in the US, so too have rates of obesity. Interesting to note in that data they present is that although meat and especially fatty meat is often vilified and touted as the cause for obesity and health problems, all animal protein (excluding eggs) has seen only a 7% increase, whereas the two foods which have seen the most dramatic rise in this 27 year period are flour (42%), carbonated drinks (218%), fruit drinks (61%), breakfast cereals (66%), and high fructose corn syrup (1240%). One issue with identifying food addiction is that, unlike with drug addictions, there is a need for food for human survival and so natural, adaptive appetitive neural pathways are present and overlap extensively with potential addiction-like neural pathways of pleasure and reward. Various studies in animals have identified neurotransmitters and neural pathways that are active in both appetite signalling and in addiction, and this makes sense as to motivate healthy, survival behaviours such as food seeking and consumption, pleasure and reward should be stimulated. (Ifland et al., 2009) propose a distinction between triggers for normal, adaptive appetite signalling and addiction-like signalling, with the first arising in response to unrefined food and the latter in response to refined food. They theorise that humans cannot become addicted to food found in nature (such as meat, fish, fruit, vegetables, beans and grains) no matter how they are prepared, but that humans can become addicted to food refined by an industrial process (such as sugar, other sweeteners, flour, salt, caffeine, and certain fats or oils). (Ifland et al., 2009) argue that, just as drugs are made from natural elements that have been refined, concentrated and consumed in much greater quantities than the small and lower concentration among the water, fibre, vitamins and minerals (cocaine from cocoa leaves, opium from poppies, ethanol from grains, fruits or potatoes, nicotine from tobacco), so too are refined food stuffs elements that do occur naturally and thus are hardwired into our brains, but when removed from their packaging of water, fibre, minerals and vitamins are consumed in much higher concentration and quantity and thus, like drugs, can become addictive.
Furthermore, with industrial refinement they can be presented in combinations with other refined food elements that can form powerful palatable and addictive effects together, that would never be possible to experience in nature. For example; soft drinks (sugar and caffeine), doughnuts (refined flour, sugar, salt, fat and sometimes caffeine from chocolate), chocolate (sugar, caffeine, fat) and French Fries (fat, salt and sometimes sugar (both from starch and from added sugar such as dextrose).
(Ifland et al., 2009) report various quotes and anecdotes from 12 participants interviewed via a clinical, targeted food based interview, based on the DSM-IV-TR criteria for substance dependence. They report direct quotes from participants that conform to each of the criteria for a substance dependence for food, providing suggestive evidence for food addiction. They admit that observational data cannot prove food addiction is a construct, but is essential in the initial hypothesis generation phase.
For the tolerance criteria, (Ifland et al., 2009)cite rat studies showing that rats allowed to self-administer, increased glucose consumption progressively, indicating tolerance to sugar. These studies also observed the neurochemical changes resulting from or causing tolerance. Excessive stimulation of neurotransmitter receptors in drug addiction results in the down-regulation of those neurotransmitters and this results in tolerance symptoms. A study in humans showed that D2 dopamine receptors were downregulated in obese individuals in a manner mirroring that of methamphetamine users (Ifland et al., 2009). The initial craving symptoms, pre-tolerance, have been seen in rats, wherein sucrose cues result in the release of dopamine and caffeine use results in the release of opiates (Ifland et al., 2009). Further rat studies show that antagonists for serotonin, cannabinoids and opioids result in the extinguishment of cocaine and sucrose seeking behaviour as well as general food seeking behaviour (Ifland et al., 2009).
For the withdrawal criterion, rat studies again have shown that rats which have become accustomed to feeding on sugar, but then are deprived exhibit shaking paws, teeth chattering, head shaking and anxiety, similar to the withdrawal symptoms rats exhibit from morphine withdrawal (Ifland et al., 2009). Some prelimary evidence for salt withdrawal in humans has been observed with symptoms of nausea, while caffeine withdrawal symptoms of headache, fatigue and drowsiness has\v been observed in humans (Ifland et al., 2009).
For the use more than intended criterion, indirect evidence comes from the rising obesity, the reports of failed diet plans, the amount of money and time spent on diets trying to control eating, the reports of the majority of dieters who lose weight regaining weight and the extensive literature on binge-eating.
For the tried to cut back or eliminate use criterion, the above statistics on the amount of money spent on diet programs and products are also indicative. Various diet studies report that people are increasingly conducting restrictive dieting practices and from earlier and earlier ages (Ifland et al., 2009).
For the spend time using, pursuing or recovering from use criterion, (Ifland et al., 2009) suggest that although directionality of cause and effect has not been established, there may be a direction opposite to that generally assumed between sedentary lifestyle and obesity. It is often assumed that sedentary lifestyle is a cause of obesity, but it may be that it is also an effect of obesity, brought on by refined foods. As consumption of refined foods increases, lethargy and drowsiness, especially with blood sugar spikes and crashes, cause an increase in TV watching and sedentary lifestyle. Further the amount of money and time spent on dieting programs and products is suggestive of time and money spent on recovery from food issues. Bulimia Nervosa (BN) and Binge-Eating Disorder (BED) literature also suggest abnormal measures taken to acquire and consume large amounts of specific craved foods in certain “comfortable” bingeing situations (avoiding socialisation). The purging effects that follow, whether by caloric restriction, fasting, induced vomiting or extreme exercise are also indicative of time spent recovering from substance abuse.
For the missing out of activities criterion, (Ifland et al., 2009) cite studies showing that obese people are more likely to avoid social situations, have less chance of pregnancy, avoid exercise, lowered chance of obtaining quality healthcare, lessened employment opportunities, and lowered rates of completing education. Further, the social stigma of obesity, and of binge-eating behaviour often results in shame, body image shame and the avoidance of social situations or other activities.
For the continued use despite knowledge of negative consequences criterion, (Ifland et al., 2009) cite studies showing the non-compliance for dietary changes in patients for cardiovascular disease, diabetes and hypertension treatments. Rising obesity rates themselves are evidence for continued use of obesity inducing food, despite knowledge that it is bad for them. Survey data show that knowledge is not the limiting factor in weight-loss, but people’s self-expressed lack of willpower is the limiting factor.
One helpful quote that describes the distinction between normal, adaptive neural reward enforcing mechanisms for food seeking and consumption, and those maladaptive for refined foods, is as follows;
‘‘Psychoactive substances.
. . disrupt the very emotions that evolved to regulate our
behavior. They arouse reward mechanism artificially, thus stimulating
the circuits that are normally fired by an events that provide
a huge gain in fitness; but they provide no fitness gain, they simply
create an illusion” [77]. Nesse RM. Evolution and addiction. Addiction (Abingdon, England)
2002;97:470–1.
In addition to substance dependence, the DSM also offers diagnostic descriptions of substance intoxication, detailing changes and disturbances in “perception, wakefulness, attention, thinking, judgment, psychomotor behavior, and interpersonal behavior”. Might these also be present in certain food consumption?
Substance abuse is also defined in the DSM as involving the failure to fulfil major role obligations, recurrent substance use in situations in which it is physically hazardous, substance related legal problems, and continued use despite having recurrent social or interpersonal problems caused by the substance. Might these also be reflect in food related issues? Would the issue of relapse also be prevalent in food use issues?
With empirical evidence for food addiction in animal models and suggestive evidence in human populations, (Gearhardt, Corbin, & Brownell, 2009b) created the Yale Food Addiction Scale. Building on mounting evidence that food, especially highly palatable food activates the same neural pathways as drug use, namely the dopamine and opiate systems, the authors note that food and psychoactive substances both have their perceived reward value enhanced by dopamine, which, while not being equivocal to being addictive is suggestive (Gearhardt et al., 2009b). Dopamine and opiate blockers can lower the reward value and craving for sweet high-fat food in normal weight and overweight binge-eaters (Gearhardt et al., 2009b). Therefore the Yale Food Addiction Scale (Gearhardt et al., 2009b) was developed as a tool to measure this area of potential addiction symptoms in individuals. It was found to have incremental validity in measuring variance in binge-eating behaviour beyond pre-existing binge-eating measures. Although the scale is a self-report survey, it can offer further data on the prevalence and nature of the subjective experience of addiction like thoughts, attitudes and behaviours regarding food.
(Corwin & Grigson, 2009) detail that the concept of food-addiction is pervasive in the lay media, yet controversial in scientific research. (Corwin & Grigson, 2009) propose that rather than high sugar or high fat food being addictive alone, these types of food can become addictive when their high palatability is combined with patterns of restriction and bingeing. Citing rat studies where rats are systematically allowed to gorge on sugar, but then are denied sugar, they propose that certain highly palatable food types are not addictive in and of themselves but cane become so when addiction-like behaviour is entrained by a series of repeated binges followed by restrictions (Corwin & Grigson, 2009).
(Davis & Carter, 2009) noted that alcohol, opium and other drugs were historically seen as moral issues when abused, as personal failings rather than any inherent addictive properties of the substances themselves, until the 1800s and then only in science in the 1900s. It was not until later, after World War II, that the nicotine in tobacco was recognised as an addictive substance. They suggest that this may also be the path of certain food substances. Now a stigma, not of moral culpability, but of mental weakness, or lack of willpower is attached to obesity, but they suggest that in coming years the addictive properties of some food may be scientifically acknowledged. (Davis & Carter, 2009) are careful to note however that not all compulsive eating, or overeating, bingeing or obesity can be due to addiction and that, if established as a viable scientific construct, it will likely be just one explanation among many.
(Davis & Carter, 2009) explain that two of the neurotransmitters most involved in substance abuse are serotonin and dopamine and that these are also essential to optimal human functioning. They theorise that in pre-agricultural times, with unreliable food sources, the levels of these neurotransmitters would have also been irregular because both are built from precursors which must be attained nutritionally, from high quality dietary sources such as protein. Depleted levels of serotonin and dopamine can negatively impact mood, cognitive abilities and motor abilities, so maintaining consistent food sources for and levels of these neurotransmitters would have been vital. Although it has been assumed that evolutionarily ancient humans would have had minimal exposure to or use for plant based psychoactive compounds, a theory by (Sullivan & Hagen, 2002) suggests that some early hunter gatherer tribes may have used these neurotoxic plants selectively. They suggest that plants such as the coca plant (from which cocaine is manufactured) may have evolved neurotoxins as a protection against herbivores, but that some humans counter-exploited these neurotoxic effects, utilising them as an easy source of calories that also had mood lifting, fatigue preventing, appetite diminishing effects that allowed them to endure the discomfort and stress of periods marked by food scarcity (Sullivan & Hagen, 2002; Sullivan, Hagen, & Hammerstein, 2008). It is further suggested, based on evidence from some hunter-gatherer tribes, that selective use of neurotoxic plants (in this case tobacco) may have been used to protect against parasites (in this case worm burden in the gut) (Roulette et al., 2014). As history and human technology progressed, humans were increasingly able to process, refine and concentrate the compounds in psychoactive plants and hence the possibility of abusing and overusing these compounds increased.
This probable history of potentially harmful plant neurotoxins in small doses, with selective adaptive use giving way to abuse and overuse with human industrialisation is particularly informative for the way that food addiction may be considered. Foods high in sugar and fat are calorically dense and release neuropeptides which enhance mood, and also reinforce their selective consumption {Davis, 2009 #1843}. The reinforcement of consumption of sweet food or food high in fat would have been evolutionarily adaptive in times of food uncertainty. The industrial revolution has seen a dramatic increase in sugar, and even sweeter sweeteners such as high fructose corn syrup, and artificial fats such as vegetable oils which take advantage of natural reinforcement circuits in the brain, which were adapted for naturally occurring tastes in much smaller concentrations, packaged with water, fibre, minerals and vitamins.
Just as drug dosage or potency have increased with greater technology, so too have portion sizes of food and concentration of fats and sweeteners and salt. The progressive increase of portion size in food is suggestive of two things; an addiction to food and attendant tolerance resulting in need for increasingly more, and an interference of some foods, especially carbohydrates and sweeteners such as fructose with normal satiety signals.
Though there are considerable overlaps seen between compulsive overeating, binge-eating disorder or bulimia nervosa, and obesity, this does not mean that the constructs are equivocal. Nor is overeating necessarily of a compulsive or addictive type, there are no doubt various scenarios or situations in which overeating may occur as a more intentional, or a more passive behaviour. As with previously discussed literature reviews, {Davis, 2009 #1843} note research showing that food behaviours can match the criteria offered by the DSM-IV-TR for substance abuse. For example, animal studies have shown that in rats, a sugar enhanced diet increases food intake over time {Davis, 2009 #1843}. Clinical evidence of this is seen in humans with binge-eating disorder, whereby the binges become larger and more severe over time and body weight progressively increases {Davis, 2009 #1843}. Further, many binge-eaters report being overweight before beginning their binge-eating behaviour, suggesting that overtime, fattening diets may prompt greater intake and eventually binge-eating behaviour {Davis, 2009 #1843}.
{Davis, 2009 #1843} also detail some of the risk factors for developing what seems to be a food addiction construct. One theorised risk factor is reward neural pathway sensitivity, dictated by the density of dopamine receptors, the quantity of dopamine released into the synapse, and the speed of re-uptake {Davis, 2009 #1843}. Reward sensitivity has been associated with both drug addiction and compulsive overeating, however the direction of reward sensitivity responsible for higher addiction or compulsion risk is debated. Some argue that hypo-dopaminergic functioning leads to increased risk for addiction because the substance abuser attempts to self-medicate in order to compensate for their sluggish, blunted dopaminergic system. On the other hand, some argue that it is hyper-dopaminergic functioning that increases risk for developing addictions, because of an enhanced, exaggerated response to any dopamine inducing behaviours or substances {Davis, 2009 #1843}. It is possible that both heightened reward-sensitivity or blunted reward-sensitivity could both confer risk of addiction or compulsive eating.
Another risk factor is heightened levels of impulsivity, which may have biological, temperament based origins. This can result in, or be coupled with poor-decision making, in choosing immediate gratification over a longer-term but more optimal reward. Part of this decision making process is based on biological, temperament based impulsivity levels (the degree to which one will automatically choose immediate rewards) but another is a learnt process, the accumulation of memories and past experiences. These accumulated memories of how similar actions and choices have worked out in the past result in a kind of gut-feeling when a similar situation arises that influences the feeling of probability of reward and outcomes. Bulimia nervosa and obese women have been observed to have decision making deficits, as with drug users, with an inability to advantageously assess future consequences of actions {Davis, 2009 #1843}.
If food addiction is a valid construct then it would have significant implications for changes in treatment. Although CBT has been successfully used in the treatment of both drug addiction and Binge-Eating Disorder {Davis, 2009 #1843} there are individuals for which it is not effective. An additional problem for food addiction is that, unlike drug addiction, addicts cannot completely abstain from food consumption. On the other hand, one danger of labelling compulsive overeating an addiction is that individuals may relinquish responsibility and believe they have no control over their eating behaviour. The simple presentation to compulsive overeaters that they are subject to a neurobiological addiction or dependence, but one that can be overcome, may provide sufferers with greater self-empathy and yet a therapeutic sense of empowerment in the knowledge of exactly what and why they are having issues. The realisation that their maladaptive neurobiological focus on food as the major or sole source of reward in life could highlight the need for finding alternative, adaptive sources of reward outside of food consumption {Davis, 2009 #1843}. It is also imperative to understand which foods are being abused and whether they can, in fact be avoided in favour of healthier and less addictive food types. Further negative emotional triggers may trigger an addictive cycle as reward is sought, meaning that addicts may need to improve emotional regulation and stress or distress tolerance.
Learned habits, associations, triggers and external stimuli may also have to be relearned or avoided. One approach that has been used with success in drug addiction therapy is CERP (Cue Exposure with Response Prevention, whereby the powerful external stimuli or cues that are associated with the unwanted substance consumption are extinguished. This may be especially useful for food addiction where external stimuli are so prevalent and difficult to avoid. In this training, the target image, smell or whatever external trigger or stimulus is repeatedly presented to the patient, but they are prevented from enacting the attached behaviour (eating that food). The prediction is that repeated use of this training will uncouple the conditioned associations and extinguish the compulsive choice to eat that food {Davis, 2009 #1843}.
A further result of recognising compulsive overeating as an addiction may be the use of similar drugs that have been used with other addictions, such as Naltrexone, which blocks opiate receptors involved in the release of dopamine and reward pathways {Davis, 2009 #1843}.
One issue that {Davis, 2009 #1843} note is that food addiction, because food and eating behaviour in general are such regular, daily and socially prevalent behaviours, straddles elements of both substance dependence/addiction and behaviour dependence/addiction.
Binge eating is usually accompanied by feelings of loss of control and guilt, shame, or psychological distress (Mathes, Brownley, Mo, & Bulik, 2009), which fit some criteria for substance dependence. Stress, negative emotions and environmental stressors have long been associated with initiation of binge-eating episodes, however they probably also contribute to enhancing the reward sensations of the binge-eating behaviour and the reinforcing efficacy {Mathes, 2009 #1842}. Food restriction or food deprivation consistently increase subsequent binge-eating behaviour in animal models and similar effects are seen in humans. When individuals have engaged in dieting or other food restriction, the risk of binge-eating in non-clinical populations is increased and the intensity and duration of binge-eating in BED or BN patients is increased {Mathes, 2009 #1842}. Attempts to emulate this in animals has shown some interesting results. When rats have their food restricted for periods but then are allowed to eat freely for subsequent periods, although their binges seem excessive when taken singularly, they tend to achieve a balance of caloric intake resulting in their body weight returning to 95%-105% of their natural weight. It is only when this pattern is paired with stressors or with highly palatable food that this restriction-refeeding pattern results in significant binge-like increases in calories in rats {Mathes, 2009 #1842}.
The reverse also seem to hold true in animal models, following a highly replicable pattern in rats {Mathes, 2009 #1842}. Stress alone also does not instigate food addiction or binge-eating based overeating, rather it leads to a caloric decrease of intake. To instigate compulsive binge-eating patterns, the stress must come after 3 cycles of restriction/refeeding and then soon after the last restriction/refeeding cycle and palatable foods must be present {Mathes, 2009 #1842}. Similar patterns are seen in the distinct ways that stressors can impact eating behaviour in humans. Though there is individual difference in results of stress, generally physical stress (such as threats of some kind of harm to the body), or stress that is indexed in physiological symptoms result in decreased caloric intake, whereas psychological stress results in increased caloric intake and increased snacking. History of dietary restraint, external eating, emotional eating, being female, obesity, general disinhibition and other individual differences can all be risk factors for developing binge-eating, but importantly the pairing of stress and cycles of dietary restriction/refeeding is a potent combination as a risk factor for developing binge-eating {Mathes, 2009 #1842}. It seems in the human data that the stressors do not increase metabolic need, but instead increase need for reward {Mathes, 2009 #1842}.
A further risk factor for “sugar-addiction” may be familial history of alcohol abuse {Fortuna, 2010 #1846}. They report that similar symptoms of craving, tolerance, withdrawal are present in drug and sugar-addiction and that cross-sensitization has been observed in sugar-addiction to narcotic dependence {Fortuna, 2010 #1846}. Research has also shown a higher rate of sucrose, or sweet-food preference among drug and alcohol dependant patients {Fortuna, 2010 #1846}. The release of dopamine in the nucleus accumbens as well asother neural pathways between sugar and drug consumption and addiction share significant overlap. They further report that the biological children of alcoholic parents, especially fathers, display higher risk of developing strong sweet food preferences and a higher risk of developing an eating disorder. Specific genes have been implicated in the development of higher sweet food preference, with some common genetic components shared by bulimia, obesity and alcohol abuse such as the A1 allele gene and the dopamine 2 receptor gene {Fortuna, 2010 #1846}.
{Benton, 2010 #1844} notes the problematic nature of defining addiction, as there are calls for addictive or compulsive behaviours such as shopping, gambling, pornography or other sexual addictions, internet use or TV watching, and other behaviours that do not involve the consumption of any compound. Further there are distinctions between physical, pharmacological dependence and psychological dependence, though they usually co-occur {Benton, 2010 #1844}. {Benton, 2010 #1844} raises an interesting objection to the conclusions drawn in some rat research on sugar-addiction, where generally the rat is given a choice between some sugar solution and rat chow, and asks whether it is the sucrose, the sweetness or simple greater palatability that causes the rats to increase consumption of the sugar. He notes studies where rats fed on sucrose that was not metabolised, or on sugars other than sucrose also elicited reward based neuronal activity {Benton, 2010 #1844}. However, this does not challenge the case for sugar-addiction so much as broaden it and highlight the need for a fine-grained research into what types and under what circumstances these symptoms may be elicited in rats.
Both addictive drugs and palatable food release dopamine in the nucleus accumbens {Benton, 2010 #1844} but it is unclear what differential effects different foods have on reward systems. It is generally theorised that, rather than food initiating similar reward pathways to drugs, drugs are actually initiating neural pathways that pre-existed their appearance and were developed for food and sex reward. This reward network, assumed to reinforce adaptive behaviours for survival, health and procreation, is complex but involves the median forebrain bundle, the nucleus accumbens, the ventral tegmental area, the ventromedial and lateral nuclei of the hypothalamus and the amygdala. {Benton, 2010 #1844} cautions that although sucrose entering the mouth elicits a dopaminergic reward system response, so too do many other pleasant experiences, such as listening to liked music, being in love, enjoying attractive or smiling faces, winning a prize or a mother seeing her child, so the context within which neural sucrose responses are observed must be remembered. It is difficult to reach any conclusions based on the dopaminergic response as it is also a normal neural response for a pleasant experience, though it does allow, rather than discount the possibility that sugar might be addictive.
Although drug use and natural food reward systems do overlap, lesion and antagonist studies have shown there are some different neuronal areas related to each {Benton, 2010 #1844}. One study showed another difference between sucrose consumption and cocaine consumption, noting that both resulted in elevated dopamine levels before consumption, but that only cocaine resulted also in dopamine release after cocaine consumption. So although dopamine quickly returned to baseline levels after sucrose consumption in rats, they underwent a second spike after cocaine consumption {Benton, 2010 #1844}. Also of note is that dopamine responses see similar elevations each time a novel palatable food is presented to a rat, but when the same food is presented second or multiple times there appears to be a dampened dopamine response, showing that food rewards seem to habituate, whereas cocaine rewards do not. In rats, regular feeding including sucrose tends to habituate whereas irregular, restriction/refeeding schedules result in continued dopamine releases {Benton, 2010 #1844}. Thus there is suggestive evidence in animal models that sucrose alone is not addictive, but that the manner in which it is consumed may be.
{Benton, 2010 #1844} further notes that though food cravings do exist, they vary considerably from that of drug addiction and that they do not conform to what would be expected by a sugar-addiction model based strictly on the cravings of a drug paradigm. He notes that a sugar-addiction hypotheses would predict that cravings should increase over time, that tolerance should result in greater intake and that periods of restricted or lowered intake should result in greater cravings, but these do not occur. Children exhibit stronger preference for sweet food but that generally diminished over the lifespan and people generally crave sugar less after periods of abstinence or lowered intake. Further he notes that cravings often occur in relation to emotional stimuli such as negative affect, stressors, feeling bored or others. He also notes that the highest cravings reported across food types are for foods high in fat (although he fails to highlight that they are also high in sugar) {Benton, 2010 #1844}. In fact, foods high in sugar alone (such as candy or sweet soft drinks) tend to be craved less, whereas foods with high amounts of both sugar and fat tend to be the most highly craved, with women tending towards a slightly higher ratio of sweetness and men toward a slightly higher ratio of fat). It is interesting to note that in nature, apart from breastmilk, there are virtually no foods that contain similarly high amounts of both sugar and fat, rather, natural, unprocessed food tends to be dominant in one or the other. So it is possible that although high sugar foods may be addictive on some level, they do not elicit cravings the way that combinations of sugar and fat do. Likewise, foods high in caffeine generally do not produce cravings alone (as is seen in the absence of coffee cravings, but the habitual consumption nonetheless, without conscious cravings), until caffeine is combined with sugar and fat in milke chocolate (highly craved), but less so in dark chocolate (less sugar and less fat, but higher caffeine).
{Benton, 2010 #1844} further reports that some studies show that fasting over 3 weeks, or dietary restrictions in obese patients over 20 weeks leads to lowered cravings rather than increased, as he suggests a sugar-addiction model would predict. However this would be only one food addiction model and does not conflict with the hypotheses of food-addiction models, but rather with his presentation of one type of food-addiction model. The lowering of cravings for certain foods over 3-20 weeks in fact aligns with other food addiction models which would suggest it is a return to more natural food appetite after weaning off cravings. A further comparison of food and drug cravings by {Benton, 2010 #1844} is problematic. He states that if food addiction is a major cause of obesity than it should dictate food intake on a regular basis. He then quotes self-report studies showing that food cravings are felt by the general population and obese populations on average between “sometimes” and “rarely”. He concludes that therefore food addiction is not a viable construct that is driving obesity because drug addicts feel cravings daily that lead them to their drug use. He says that since we eat several times a day one would expect any food cravings to be experienced multiple times daily. The first problem with this argument is the equation of food-addiction with food cravings. This is misleading in equating the two. Food addiction, if it exists, may include food cravings as a key feature but not be equated with regular food cravings. Further, drug use is often defined especially by cravings only when the drug is abstained from or resisted, as per the DSM criteria for substance dependence. In typical drug use, if the habitual, compulsive abuse of the substance is regularly continued, consciously experienced cravings would be less frequent. This may be the case with food addiction too. If the target food of abuse is consistently consumed and not resisted or abstained from, then consciously experienced cravings would not be expected as frequent occurrences.
The point he makes regarding the lowering of cravings after abstinence is interesting though and does warrant further examination. If food addiction is a valid construct and the cravings operate similarly to that of drugs, then it would be expected that abstinence would result in initial cravings which would intensify. It would also be predicted that given long enough abstinence, these cravings should reduce in frequency and intensity. Although he reports the lowering of cravings after prolonged absence as indicative of the invalidity of a food addiction model, it may in fact support such a model. Until evidence shows a curve of food craving experience over time, it is premature to present this as either supporting or challenging a food addiction model.
He also proposes that because chocolate craving is experienced differentially across the menstrual cycle that it differs from drug dependence, which is minimally affected by menstrual cycles. He also notes that both alcohol and nicotine dependence have cravings from first thing in the morning, presumably because of an overnight fast, whereas high sugar cravings or other food cravings are often experienced later in the evening. From this he concludes that, because cravings patterns are different, that a food addiction model is unreasonable. Though his conclusion is far from warranted it is an interesting avenue for further exploration, the differential daily craving cycles of carbohydrates and why they tend to occur later in the evening.
One other challenge that {Benton, 2010 #1844} suggests is that if sucrose is addictive, then a tolerance (and hence greater need for greater quantities of it over time) should be observable. Yet generally, humans have a taste preference for very sweet food when younger that diminishes to a preference for moderate sweetness in adulthood. This is an interesting challenge that requires further research. However, he fails to address data in eating disordered populations or obese populations that indeed show progressively increasing consumption of sweet and sweet/fatty food and drinks over time.
As {Benton, 2010 #1844} notes, there is a distinct difference between human homeostatic eating, and hedonic eating. It is also important when considering neural changes and compulsive behaviour to recognise that hedonic eating does not equal compulsive or addictive eating, but that much hedonic eating is driven by palatability, texture, smell and other external stimuli, whether latent or conditioned associations {Benton, 2010 #1844}. Though it is true that much hedonic consumption is different from addiction, the fact that it occurs does not discount the existence of food addiction with some foods, in some individuals and some circumstances, but it does highlight the need for care and untangling of constructs with careful research design and conclusions drawn only on the evidence pertaining to what is observed in that research paradigm, rather than generalised to a broader population.
{Benton, 2010 #1844} then goes on to suggest that in fact sugar is not responsible for obesity and that the fat and water content of food are the primary drivers of obesity, along with energy density of food. He cites a handful of articles from the 1970s and a couple from the 1990s, with only one cited within 10 years of his publication (which is itself based on older research from the 1990s) to support his implication that it is dietary fat, not dietary sugar that is responsible for the obesity epidemic. Unfortunately he ignores all the other data that was available at the time, and which has grown in strength since, suggesting that added sugar consumption (whether sucrose or otherwise) has increased along with obesity and that fat intake has in fact decreased. The ignorance of evidence that points in the other direction and the careful selection of phrasing in this passage looks almost as though the author is attempting to obfuscate rather than elucidate possible causes for obesity.
{Benton, 2010 #1844} raises some valid concerns and challenges to the validity of a sugar addiction construct by referencing research which shows that Binge-Eating Disorder populations show binge-eating behaviours for a wide variety of food types, not only highly sweet only food. He also notes that there are a wide range of possible causes for BED, many of which are social, environmental, genetic and psychological, rather than nutritional. However, this has never really been a cornerstone of any theory proposing food addiction of any kind, that nutrition alone is a causal trigger in BED or BN. Further, if nutrition is not a main initial cause of BED or BN it does not preclude any food-addiction models. He then goes on to wilfully simplify the possible aetiology of BED, setting up an either/or proposal; that binge-eating is other directly from psychological issues or directly from the diet consumed. However, given the wealth of evidence that it is likely a complex and variable combination of various factors, presenting this two options is a wilful simplification, designed as a setting up of a straw-man argument.
In examining the types of food that is typical consumed in binge eating episodes {Benton, 2010 #1844} notes that the data suggests a variety of foods that are not marked by solely high sucrose foods. This is correct and does suggest that a food-addiction model which references binge-eating and is focussed solely on sucrose will be problematic. However the range of foods does reveal a highly dominant preference for other sugars and carbohydrate rich foods, as well as a decrease in protein and increase in other processed, refined foods. This suggests that natural sources of fat and protein (meat) are not generally preferentially consumed in binges, but that processed versions along with a dominance of refined carbohydrates and sugars and processed fats are. Therefore, his explanation does not necessarily shift the target from sucrose but widens the target.
{Benton, 2010 #1844} concludes with reasonable recommendations that the evidence of opioid blockers has mixed results for a food addiction model, that rat based research cannot be generalised to human populations and that conclusions should not be hastily drawn based on current inconclusive research. He shifts the target to fat in general, although it should be more accurately placed on processed or refined fats as these are the substances of binges, along with other processed carbohydrates, salts and sugars. He also closes by acknowledging that his research was partially funded by World Sugar Research Organization, though expresses he has no financial interest in the sale of sugar or sugar products.
{N Gearhardt, 2011 #1829} analysed a population of obese patients with Binge-Eating Disorder using their Yale Food Addiction Scale. They noted that of those surveyed, 57% fit the YFAS definition of food-addiction, although because they are defined that only by that scale, the authors are careful to use the term “food-addiction” in quotation marks. The YFAS is based on clinical models of drug addiction and DSM-IV-TR criteria for substance dependence, with items describing those symptoms in food consumption terms. Food-addiction as defined by the YFAScale appears to overlap significantly with Binge-Eating Disorder but to be differentiable. Those BED patients who met the threshold for Yale Food Addiction Scale based “food-addiction” seemed to be defined by greater impulsivity, greater problems with emotional regulation, psychopathology and negative affect. The results indicate that in Binge-Eating populations, food-addiction like symptoms are largely centred on highly processed foods that are high in sugars and fats {Gearhardt, 2012 #1835}.
{N Gearhardt, 2011 #1829} wrote that food addiction, although a commonly used term in public media for years has only become credible in scientific circles with increased neural evidence and a loosening on the restrictions of what constitutes addiction. They present a list of similarities between hyperpalatable foods and addictive drugs, including;
Activate dopamine and opioid neural circuitry (but how are some food different from others in their activations of these?)
Trigger artificially elevated levels of reward (but again, how much higher is this than in normal non-addictive but pleasant food?)
Absorbed rapidly into the bloodstream.
Alter neurobiological systems
Cause compensatory mechanisms that result in tolerance
Combined with additives to enhance rewarding properties
Elicit cue-triggered cravings
Consumed in spite of negative consequences
Consumed in spite of a desire to cut down
Impact disadvantaged groups to a disproportionate degree
Cause high public health costs
Exposure in utero can result in long-term alterations
The industrial processing of what were previously natural food compounds has led to drugs and hyperpalatable, concentrated and rapidly absorbed foods. For example, the leaves of the coca plant, when chewed or stewed as tea, it produces only mild stimulation with little addictive potential. However, when industrially processed and concentrated it becomes hedonically rewarding, rapidly absorbed and highly addictive {N Gearhardt, 2011 #1829}. A similar difference between naturally occurring food and highly processed food, with a prime example being high fructose corn syrup. Like high fructose corn syrup, ethanol (alcohol) is simply the fermented byproduct of fructose, from ripe fruit. Both may, in small doses provide minimal caloric value but in large doses can be devastating for health. Fructose may be especially damaging to appetite regulation as it promotes insulin resistance and increased release of triglycerides in the liver {N Gearhardt, 2011 #1829}. Fructose appears to have different triglyceride raising properties than other sugars, blunts leptin (and hence satiety) responses and thus increase hunger and promotes greater cravings separate from energy needs, and as mentioned, unlike glucose, does not stimulate the release of insulin to such a degree {N Gearhardt, 2011 #1829}. Many processed foods result in disrupted blood glucose levels, which result in increased insulin responses and a cascade of other disrupted physiological adaptations to deal with the dysregulated homeostasis.
{N Gearhardt, 2011 #1829} presented a brief history of addictive substances and highlighted that many of the substances that we take as self-evidently addictive, such as cocaine, nicotine, alcohol and heroine, were more commonly accepted until the 19th century when, at various times, concern began to swell over them. Another issue to be cognizant of is that not all addictive drugs exhibit every facet of addiction. For example, cocaine exhibits almost no physical withdrawal symptoms so that it was controversial to call it addictive until the late 1980s. Similarly, because tobacco does not involve any strong intoxication effects, it was controversial to call it an addictive substance for some time {N Gearhardt, 2011 #1829}.
(Hasin et al., 2013) DSM V changes and rationales
If food can be addictive it is important to research what kinds of food, in what quantities and for what person in what situation? So far the most studied and most likely candidate for food addiction is sugar. There is also suggestion that among types of sugar, high fructose corn syrup may be the most addictive, because of its greater sweetness and the way in which fructose can bypass satiety signals. Fat is also a food commonly cited as potentially addictive, with high sugar-high fat combinations being perhaps especially addictive, however there is little research in any potentially addiction-like effects of fat, or on which types of fat, or in what combinations?
If food addiction, in whatever form, does become a scientifically evident construct, the impact on personal lives and the food industry would be substantial. Prevention and treatment for eating disorders, obesity and other diet related mental or physical disorders would need reevaluation and possible reshaping. Concepts of company, governmental and personal responsibility would shift. Labelling laws, food industry guidelines and ingredient prohibitions may eventually shift and cause large changes to companies and connected economic concerns. Regulation, legalisation and litigation would all see shifts.
As (Gearhardt et al., 2009a) reflect, if food-addiction is found to be a viable construct, there are lessons to be learned from the alcohol and tobacco industries. Although the recognition of alcoholism as a kind of addictive substance abuse has relieved some of the stigma from alcoholics, because it can be consumed without issues of dependence by many in the population has led to a lack of responsibility from alcohol companies, sellers and marketers, with a less dramatic shift to the alcohol environment than with nicotine/tobacco. The authors (Gearhardt et al., 2009a) suggest that the failure in the US to take a public health approach to alcohol may have allowed the continued rates of alcoholism. In contrast, nicotine has been recognised as a highly addictive drug which will result in dependence for most that use tobacco products. The public health campaign in the US, Australia and other Western countries has been much stronger and the taxes levied on the industry, as well as strict regulation on advertising, marketing and product packaging, including health warnings, has seen progressive drops in the rates of tobacco and nicotine use (Gearhardt et al., 2009a). Whether the food industry followed the example of the alcohol industry (which would be likely) and simply claim that it is an individual’s responsibility because the dependence is perhaps milder and may not affect the majority of the population, or the way of the nicotine/tobacco industry with dramatic and far reaching economic implications may depend on the strength and extent of scientific research.
The issue with potential food addiction is the ubiquitous nature of most food types, especially sugar in food, to the extent that to create an environment that has restricted external stimuli as craving cues, regulation on food production, marketing and labelling would mean a monumental shift in entire industries and cornerstones of economy. These would present significant barriers to any large governmental, legal or educational shifts even if compelling scientific evidence were presented, at least in the immediate future.
Sugar and fat usually occur separately in nature. Things that are high in sugar (like fruit or honey) are generally low in fat. Things that are high in fat (like meat) are generally low in sugar. The only place that pre-agricultural humans could have gotten a significant combination of fat and sugar was in mother’s breast milk.
(N Gearhardt, Davis, Kuschner, & D Brownell, 2011)
Can too much a Self-control be Bad?
With so much in early childhood, adolescence and adulthood being benefited by greater levels of self-control and greater self-regulatory strategies, or self-discipline (the trait self-control probably measured by the (Brief) Self-control Scale) it seems obvious that the best thing to do is to build as strong and constant self-control as possible. After all if we became able to constantly delay gratification, act perfectly according to our deliberative, considered will and long-term plans for health, financial well-being and reasoned careful communication, surely our lives would be near perfect…..right? Well, it seems as though there are some glimpses of when self-control or self-discipline can be imbalanced as quiet voices of new research warn, often lost among the loud voices exalting self-control above all else.
For example, as seen above in child development, impulsivity can be considered as one facet of self-regulation, regulating the self to approach rather than to avoid, to seek excitatory stimuli and to experience new challenges and opportunities for learning. In children, an excess of impulsivity and lack of effortful control can lead to problematic externalising behaviour such as aggression or self-destructive behaviour. But on the other hand, an excess of effortful control and a lack of impulsivity can result in the child being overly cautious and avoidant, denying many approach behaviours that could be beneficial and hence leading to problematic internalising behaviours such as depressive symptomatology or anxiety or fearfulness. (see above).
In adults, (Kivetz & Keinan, 2006) found that at times, in certain situations, constantly making apparently virtuous decisions (such as choosing to work rather than indulge in pleasure) can result in the avoidance of guilt arising from indulgence, but can build feelings of wistful regret from missing out on present enjoyment. They term this excessive focus on the long-term results, and delayed gratification or deferred reward as hyperopia. However, the focus of much research into self-control is on people’s myopia, or shortsightedness, or excessive impulsiveness and consequently problematic behaviour. Certainly impulsive eating behaviours, compulsive spending, alcohol and substance abuse, uncontrolled emotion and other problems have been shown to be prevalent and to cause regret and guilt after the initial gratification dissipates. However, researchers (Kivetz & Keinan, 2006) proposed that overcontrol and excessive farsightedness can have negative long term consequences, with the continued enaction of long-term focussed “virtue” over “vice” creating an increasing regret over time.
In their paper, (Kivetz & Keinan, 2006) argue that although there may be self-control situations, where there is a clear unhealthy or destructive impulsive temptation and a clear virtuous choice for delayed gratification, there are also sometimes genuine self-control dilemmas, in which the correct choice is not immediately clear. In such situations, sometimes avoiding immediate indulgence regret could lead to building future regret of missing out, and so the best option to minimise regret may be to indulge in immediate gratification.
One approach for consumers that feel the regret of missing out from hyperopic saving, or buying restraint, is to precommit to future indulgence, to have a planned indulgence at some future time. The studies by (Kivetz & Keinan, 2006) showed that this missing out regret could occur in the arena of working, rather than indulging in pleasure, as well as studying, rather than relaxing, and eating a healthy salad, rather than indulging in a piece of cake. The further in the future the long-term reward for each self-control dilemma, the greater the level of regret about not indulging was. Further, it was found that over time, the feeling of regret in missing out could outweigh the regret of indulgence, because guilt of giving in to temptation is generally expressed as a “hot” emotion that rises and passes, whereas the regret of missing out is a “colder” affect, building gradually. This conclusionis unjustified however, as it discounts the accumulation of multiple, fleeting “hot” emotions of guilt into a more long-term feeling of shame. However, the study also found that hen cognitive processing was primed, rather than affective, this level of regret diminished. So, although the conclusions reached by the authors are perhaps exaggerated and do not consider accumulated or long-term feelings, they do highlight the danger of excess in both directions of self-control.
Also talk about perfectionism
{Berkman, 2016 #1770} use the term self-regulation in a manner that is synonymous with self-control, after establishing a definition of a goal. They define goal as a cognitive construct that s generally more broad and longer-term than more immediate hedonic goals. Then based on this
(Treglown, Palaiou, Zarola, & Furnham, 2016)
Trait Self-Control Predicts Performance on Behavioral
Tests of Self-Control
High Self-Control Predicts Good Adjustment,
Less Pathology, Better Grades, and
Interpersonal Success
Everyday Temptations:
An Experience Sampling Study of Desire, Conflict, and Self-Control
Trait self-control and the avoidance of temptation
(Wojdylo et al., 2017) (Dewitte & Lens, 1999) (Horstkötter, 2015)
Decomposing self-regulation and self-control: The Volitional Components Inventory.
J Kuhl, A Fuhrmann – 1998 – psycnet.apa.org
Abstract 1.
Introduces a questionnaire, the Volitional Components Inventory (VCI), based on
a theory of volition. This theory postulates 2 different modes of volition: self–control and self–
regulation. According to the central assumption
underlying the questionnaire, volitional
Self-dsicipline – trait?
Inhibitory control – facet of self-control? (see this on brain regions) https://www.psychologytoday.com/blog/the-athletes-way/201601/your-brain-can-be-trained-self-regulate-negative-thinking
(Tangney et al., 2004) self-control scale, various benefits
Allport, G. W. (1937). Personality: Holt New York.
Bernier, A., Carlson, S. M., & Whipple, N. (2010). From external regulation to self‐regulation: Early parenting precursors of young children’s executive functioning. Child development, 81(1), 326-339.
Dewitte, S., & Lens, W. (1999). Volition: Use with measure. Learning and individual differences, 11(3), 321-333. doi: http://dx.doi.org/10.1016/S1041-6080(99)80006-5
Freud, S. (1961). The ego and the id.
Freud, S., & Strachey, J. (1962). The ego and the id: WW Norton & Company.
Hay, C., & Forrest, W. (2006). The development of self‐control: Examining self‐control theory’s stability thesis. Criminology, 44(4), 739-774.
Karreman, A., Van Tuijl, C., van Aken, M. A., & Deković, M. (2006). Parenting and self‐regulation in preschoolers: a meta‐analysis. Infant and Child Development, 15(6), 561-579.
Kochanska, G., Coy, K. C., & Murray, K. T. (2001). The development of self‐regulation in the first four years of life. Child development, 72(4), 1091-1111.
Kochanska, G., Philibert, R. A., & Barry, R. A. (2009). Interplay of genes and early mother–child relationship in the development of self‐regulation from toddler to preschool age. Journal of Child Psychology and Psychiatry, 50(11), 1331-1338.
Pelchat, M. L. (2009). Food addiction in humans. The Journal of nutrition, 139(3), 620-622.
Schmeichel, B. J., & Zell, A. (2007). Trait Self‐Control Predicts Performance on Behavioral Tests of Self‐Control. Journal of personality, 75(4), 743-756.
Sullivan, R. J., & Hagen, E. H. (2002). Psychotropic substance‐seeking: evolutionary pathology or adaptation? Addiction, 97(4), 389-400.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self‐control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of personality, 72(2), 271-324.
Wurtman, R. J., & Wurtman, J. J. (1989). Carbohydrates and depression. Sci Am, 260(1), 68-75.
Wurtman, R. J., & Wurtman, J. J. (1995). Brain serotonin, carbohydrate‐craving, obesity and depression. Obesity, 3(S4).