Conceptual Foundations of Self-Control as Delay of Gratification

Contents

1 Introduction
1.1 Preliminary Definitions and Assumptions
1.1.1 A More General Limitation: Self-Control vs. Self-Regulation
1.1.2 Narrowing down the Definition of Self-Control
1.1.3 Further Definitions and Assumptions
1.2 What this Thesis is not about
1.3 Aims and Course of Action
1.3.1 Aims
1.3.2 Course of Action
1.3.3 Challenges
1.4 Methodology: It’s all about Distinctions
1.5 Relevance of Self-Control

2 Self-Control as Delay of Gratification
2.1 Setting the Stage - Initial Conditions
2.2 Short-Term Value Attribution as a Normative Problem
2.3 The Relation of Intentions and Behavioral Decisions
2.4 “Genuine” Delay of Gratification and the “Surrender Scenario”
2.5 Potential Moderating Variables
2.6 Provisional Summary

3 Putting it all together: An Analytical Framework

4 Summary and Future Perspectives
4.1 Summary: What Has Been Reached
4.2 Perspectives: What May Follow Next

References

Appendix: Illustrations

A Sect. 1: Introduction

? Sect. 2: Self-Control as Delay of Gratification

C Sect. 4: Putting It All Together: An Analytical Framework

1 Introduction

If you ever felt an urge to do something forbidden, resisted a while, and failed, read on.

If not, you may not be of this world.

- Hofmann et al. (2009)

Did you have a New Year’s resolution? What was it - loosing weight? Being more friendly to a colleague? Quitting to smoke? Maybe stopping to procrastinate the tasks at hand? And if that was the ease: How successful have you been in prevailing your resolutions? Whatever it was, chances arc good that by now you have had a relapse. But why is that so - that we arc apparently not able to stick to (at least some of) our good “intentions”¹ ? Obviously, this thesis will not be able to answer this question thoroughly. However, it docs intend to provide a basis for a better understanding of this question.

In folk psychology, the challenge of resisting a certain class of desires and the corresponding be­haviors (like smoking) is often referred to with expressions like “resisting temptations”, “ exercising willpower”, or “self-discipline” . And as it is often the ease with terms or phrases that arc widely used in everyday life, there is a variety of different concepts that underlie the same term, thus making it hard to efficiently communicate about it. In example, while Peter might loud himself of his extraordinary self-control, for he considers it a heroic act of willpower to stay up an entire weekend to finish an essay that is due the next week, his friend might reproach him and argue that it has been his very lack of self-control that has brought him in the situation of having to “pull an all-nighter”. Who is right then and why?

In the ease of self-control, the problem of divergence of definitions characterizes wide parts of the scientific literature, which applies many different notions of ”sclf-control-definitions (Bockarts et al., 2000). This might be at least one of the reasons why there has not yet been a unified theory of self-control that successfully integrates the more than 60 different models that can be found only in the two “Handbooks of Self-Regulation” currently available². In order to tackle this problem, it is the aim of this thesis to contribute to the clarification of the different notions surrounding self-control. Taking two “standard notions” of self-control as starting points, the following aims will be pursued:

1. Explicating the underlying conditions needed to “attest” someone a failure of self-control³ and the different notions of self-control that accrue from these conditions.
2. Proposing an analytical framework that makes different “self-control scenarios” (i.e., theories, experiments, and everyday-situations) comparable on the basis of these conditions.

A more detailed plan of action will be provided in section 1.3, while an explanation of the exact methodology used here will be exemplified in subsection 1.4. However, in order to provide readers who arc novices to the topic of self-control with a basic understanding of the vocabulary in this field of research, the following subsections will first introduce relevant definitions and narrow down the topics which arc to be discussed and which arc to be excluded. This is deemed necessary by the author to provide a sufficient conceptual foundation and a satisfying degree of clarity for the following discussion. Therefore, the preliminary course of action goes as follows: The following subsection 1.1 will provide preliminary definitions of self-control (and therefore implicitly of self­control-failure, although this will be explicated at a later point): First, by providing a general limitation or distinction between self-regulation and self-control (1.1.1); secondly, by stating and briefly analyzing a “narrow” definition of self-control, which will serve as a basis for the subseciuent discussion (1.1.2) and thirdly by providing further definitions of concepts that have traditionally been related to self-control (1.1.3), i.e. motivation and volition. Subseciuentlv, subsection 1.2 will briefly state what this thesis is not about, as to clarify towards which ciuestions and discussions it will remain neutral (although they might be interesting), most prominently the freedom of will. This is considered early enough to give those exclusively interested in these ciuestions a fair chance to decide whether they would like to continue reading. Subsection 1.3 then will finally outline the further discussions by first picking up and explaining the research ciuestions stated above (1.3.1), providing the reader with final directions of the remaining course of action (1.3.2), and explaining challenges and limitations of the present account (1.3.3).

1.1 Preliminary Definitions and Assumptions

In order to contain the field of self-regulation that is going to be assessed here, first a more general limitation of the topic at hand by Carver and Scheier (Carver and Scheier, 2011)will be considered. Secondly, an additional definition by Baumeister et al. (2007) will be taken as a starting point for the subseciuent reflections.

1.1.1 A More General Limitation: Self-Control vs. Self-Regulation

Many authors in the literature use the terms “self-control” and “self-regulation” interchangeably (i.e. Bauer and Baumeister, 2011). In their introductory chapter of the “Handbook of Self-Regulation”, however, Carver and Scheier make the following distinction:

“This term [self-regulation], however, means different things to different people. Many authors in this book use this term as roughly equivalent to self-control: Overriding of one action tendency in order to attain another goal. We prefer to reserve the term self-control for such cases and use the term self-regulation more broadly. When we use the term self-regulation, we intend to convey the sense of purposive processes, the sense that self-corrective adjustments are taking place as needed to stay on track for the purpose being served (whether this entails overriding another impulse or simply reacting to perturbations from other sources), and the sense that the corrective adjustments originate within the person.” (Carver and Scheier, 2011, p. 3)

Let us briefly explicate what the different notions of self-control and self-regulation mean. At this point, self-control is defined in a rather frugal way that roughly corresponds to one of the “narrow definitions” that will be assessed in the following subsubsection, that is to say as the ability to “override on action tendency in order to attain another goal”. The “broader” sense of self-regulation then seems to entail this ability but additionally covers any “self-corrective adjustments” taking place. It seems what Carver and Scheier mean by that is, in addition to overriding an action tendency or impulse, any adjustment that might be necessary to “stay on track of the purpose being served”. A sense of “sclf-corrcctivc adjustment” can be illustrated by the so-called “Wisconsin Card­Sorting Test” (WCST)⁴ that has been used especially on patients with frontal lobe damage . Here, the subject is confronted with a variety of cards that she needs to match using an unknown rule in order to attain certain rewards. After “discovering” such a rule of matching (i.e. by color), and applying it for a number of rounds, presumably a habit of responding in this way forms. However, after a while, the “rule” to follow changes, thus requiring the subject to adapt to the new situation by consciously modifying her responding behavior. Formally, such an adaption has been described by a more or less metaphoric reference to feedback-loops as used in robotic control. The basic­idea of this concept is that an organism (or computer controlled machine) undertakes certain steps toward a desired state, i.e. matching cards or symbols. After each step, it receives a feedback on whether it has approached the final state or moved away from it, and hence evaluates the previous action or processing step. Depending on the outcome of the evaluation, the organism will cither continue the line of action (in ease of a successful attempt to approach the desired state) or modify its behavior (in ease it has been unsuccessful) by searching for alternative means to approach the desired outcome (Misehcl and Avduk, 2011; Carver and Sehcicr, 2011). To think of goal-striving behavior as controlled by some sort of feedback-mechanism seems to correspond well to the sense of “sclf-corrcctivc adjustments” mentioned above.⁵

It should be clear that a whole set of capacities and abilities which arc in principle distinguishable from another is necessary for a person or organism to initiate sclf-corrcctivc adjustments. However, this cannot be further elaborated upon without utterly exceeding the dimensions of this paper. What should be kept in mind is merely the difference between a larger set of capabilities referred to here as “self-regulation” that entails, next to “genuine” self-control, a variety of other functions. This thesis adopts this “broader” sense of self-regulation as proposed by Carver and Sehcicr and will thus strive to examine “self-control” in its more narrow sense. Consequently a further assessment of this notion of self-control, will be presented in the following subsubsection in order to clarify the implications of this definition.

1.1.2 Narrowing down the Definition of Self-Control

In order to explicate the more narrow notions of self-control that will serve as a starting point for the subsequent analysis, it should prove helpful to deploy yet another definition:

“Self-control or self-regulation (terms that we use interchangeably) is defined as the capacity to override natural and automatic tendencies, desires, or behaviours; to pursue long-term goals, even at the expense of short-term attractions; and to follow socially prescribed norms and rules. In other words, self-regulation is the capacity to alter the self’s responses to achieve a desired state or outcome that otherwise would not arise naturally. Thus, the goal of self-control is to interrupt the self’s tendency to operate on automatic pilot and to streer behavior consciously in a desired direction” (Bauer and Baumeister, 2011, p. 65)

This definition obviously entails a variety of components that require some clarification. A dis­tinction between “self-control” and “self-regulation” has already been provided and will therefore be used in the sense outlined in the previous subsubsection (and not interchangeably). This also affects the notion “self-regulation is the capacity to alter the self’s responses...”: According to the general limitation of the term “self-control” as made above, this alteration of responses entails a variety of capacities that arc not considered to be part of “genuine” self-control as it will be used here.

Here, I will argue that the Baumcistcr-dcfinition of self-control includes - in exclusion of the notion of self-corrective adjustments mentioned above - two separate senses or conditions of self-control that can be explicated. First, “the capacity...to pursue long-term, goals, even at the expense of short-term attractions” and secondly, “the eapaeity to override natural and automatic tendencies” (which is synonymous to ’’interrupt the self’s tendency to operate on automatic pilot...”) .

The first notion in which self-control is defined here is simply stated as “the eapaeity to pursue long­term goals, even at the expense of short-term attractions”. We will concentrate on the combination of long-term and short-term attractions. This setting, where the long-term goal typically represents something more valuable or desirable and the pursue of one of the goals excludes the pursue of the other has been termed “delay of gratification” (i.c., Misehcl et al., 1972, 1989, from here on: "DoG"). For this setting, as it is introduced in the literature, usually involves some implicit assumptions to be used in its narrow sense, we will preliminarily term it “two-goal conflict” and make this the starting point of the analysis of the conditions and different scenarios involved in such a setting (including “genuine” DoG) in section 2.

The second sense of self-control will be summarized as “effortful self-control” or “effortful inhibition” for it implicitly entails a distinction between “automatic” and “controlled” behavior. Accordingly, an automatic behavior (or tendency or desire) is something that arises “naturally” insofar as it docs not require any conscious thought or planning, whereas the eapaeity to override such a behavior is “controlled” or “conscious” insofar as it requires some sort of mental effort. As an illustration of these types of behaviors might serve the different states a person is typically in when learning a foreign language as opposed to speaking one’s native language. While the generation of speech in our native tongue is typically something that docs not require us to consciously select adequate vocabulary or grammatical structures, it is often necessary when we try to express ourselves in a language that we arc novices to. This corresponds to an assumption underlying a large amount of the literature on self-control, namely the functional distinction of two neural networks or systems - one primarily associated with automatic perception and behavior and one with rather conscious or reflective behavior that is used to “override” (e.g. inhibit) the other system (for further reading see: Hofmann et ak, 2009, 2011; Masieampo and Baumeister, 2008; Strack and Deutsch, 2004; Brass and Haggard, 2007)⁶.

It seems noteworthy that the Baumcistcr-dcfinition explicitly states the two notions of self-control separately, while Carver and Sehcicr’s definition states them in a united sentence: “Overriding of one action tendency in order to attain another goal”. In this paper, these notions will be treated separately. In fact, an assumption underlying this analysis is that the two notions of self-control explicated above refer to two (principally) distinguishable sets of events that do overlap in particular eases (e.g., two-goal-eonfliets that arc solved using effortful inhibition) but not necessarily so (figure 1). However, we will only be able to assess the DoG paradigm here in order to show that it should be divided into certain subsets (“scenarios”) that need to be considered in order to gain a high degree of conceptual clarity⁷.

1.1.3 Further Definitions and Assumptions

There arc a number of terms and definitions in psychological research and discourse that at least partially resemble the definitions above and hence touch the scope of this thesis. Since it is one of the aims of this work to provide a basis for unified nomenclature in self-control research, it obviously bases on the assumption that the vocabulary use of the related concepts is somewhat arbitrary because it is used differently by researchers and additionally confuted by everyday use. Hence, it docs not, strictly speaking, serve the present undertaking to provide a great number of definitions that arc actually deemed insufficient to enable a productive discourse on self-control. However, the reason to briefly list the related concepts is twofold. First, it raises the novice reader’s awareness of the related vocabulary and further reading. Secondly, it clarifies the use of the terminology in this paper, thus avoiding confusion about the scope of the present work. The last two terms, namely goal and the domains of self-control, resemble relevant assumptions underlying the further analysis of self-control.

Volition, often defined as “the capability to pursue one’s will”, is obviously concerned with the assessment of the ability to act planfullv and to “exert self-control” and with the failure to do so. However, in order to maintain a certain terminological clarity, volition will be regarded as an umbrella term for the entire process of establishing long-term goals, planning of a course of action, executing it while shielding it from interfering stimuli and ultimately evaluating the action. Thus, this paper will stick roughly to a conceptualiza­tion of volition that corresponds to the venerable “Rubicon Model of Action Phases” (i.e. Hcekhauscn et ah, 1987)⁸, while being fully aware that this term, too, has been defined in different ways in other sources (one being a more “philosophical” conceptualization as “an act of will as distinguished from the physical movement it intends to bring about”, according to the Collins English Dictionary⁹ ). However, if the term “volition” is being used in a different way than the one outlined here, the respective notion will be explicitly stated.

Impulse Control, which has been an important keyword in recent research related to self-control (i.e. Hofmann et ah, 2009; Fujita, 2011) will be, although a variety of slightly different notions exist in the literature, treated as denoting the same classes of events as the “narrow definitions” of self-control described above (subsection 1.1.2) and will thus be used synonymously. Again, the term might be thought of as only describing a certain class of self-control, namely the notion of “effortful inhibition”, as impulses arc typically thought of as occurring automatically. But since this is not necessarily the only way to think of impulse control, it seems modest to keep things as simple as possible and not to make a difference in the terms being used here.

Cognitive control has been defined as “the ability to guide and control behavior and optimize out­comes” by Rougicr et al. (2005) and Millncr and Cohen (2001). Also Allport (1987) uses the following definition: “When individuals perform tasks in the service of a desired goal, cognitive control enables them to suppress attention and responses to irrelevant infor­mation, even when the information is highly salient“. These definitions seem to roughly correspond to the additional capabilities cited in the Carver and Sehcicr-dcfinition of self-regulation as opposed to self-control cited above (subsubsection 1.1.1). However, “optimizing outcomes” as well as “suppressing ...responses to irrelevant information” for “tasks in the service of a desired goal” might as well include the concept of delay of gratification and may or may not include the notion of “consciously” overriding action tendencies. This again nicely illustrates the potential overlap of different concepts which motivated the present work. Therefore, the term “cognitive control” will not be used in this paper in order to avoid confusion.

Motivation will be used in accordance with its Latin root “movere” (to move) and therefore as a term depicting any goal or aim (on any level of representation) or concrete appetitive or aversive stimulus that is fitted to “move” a particular organism, that is to say to engage in approaching or avoidance behavior . In tandem with this definition, it seems worth mentioning that “gratification” as in “Delay of Gratification” is to be considered as denoting both approaching a stimulus or state that a subject finds appetitive or desirable (thus perceiving it as “reward”)¹⁰ and conversely avoiding a stimulus or state that a particular subject finds aversive (and perceives as “punishment”), as exhaustively instrumentalized in conditioning paradigms.

Domains of Self-Control depict the various fields in both everyday life and experimental designs that arc affected by self-control. While everyday-life conflicts presumably reflecting self-control (failure) situations have been assessed by relatively few studies (Hofmann et ah, 2012), there is a vast amount of literature on different sorts of self-control failure. There arc two important assumptions with regards to the different domains of self­control that underlie this thesis. First, what is being regarded as tasks or conflicts requiring self-control is as variant as human goals (sec below) and depends both on an individual value attribution of what goals arc worth pursuing as well as individual difficulties in pursuing them¹¹. Secondly, albeit different people might need to exert self-control in completely different domains, the underlying “mechanism”, i.e., when resisting a temptation or effortfully “forcing” oneself to pursue certain goals is basically the same across all domains. Consequently, it will also be assumed that if someone shows different “success rates” in different domains, that can primarily be attributed to one’s individual “desire strengths” in various domains and not with distinct “sorts” of self­control mechanisms that are used in particular domains (e.g. a “smoking self-control”, an “alcohol self-control” and so on.)¹². These domains include, but are not limited to, use of recreational drugs, procrastination, sports, eating, drinking, sexual desires, sleeping, social conduct (Denson et ah, 2012), purchases (Knutson et ah, 2007), and even (unwanted) thoughts - in short, any realm of life that people might feel conflicted about (Hofmann et ah, 2012)¹³.

The regulation of emotion, however, will preliminarily be considered an exception to this more or less universal application of self-control mechanisms as things might be more complicated in this domain, and a discussion of this topic would exceed the boundaries of the analysis (for a discussion, see i.e. Ochsner and Gross, 2005; Northoff, 2005)¹⁴.

Goal can be thought of as an umbrella term including one’s motives, desires, and interests and can be roughly defined as any state, behavior, object or substance a person feels drawn to (e.g. motivated by), and therefore as equivalently universal as domains of self­control. However, it is important to note that the construct of “goal”, as it is deployed here, implies a sort of “hierarchy” of abstraction. This means, that goals can be either very immanent or concrete, while being very case sensitive (such goals are often referred to as “impulses”, for instance the desire to eat the chocolate cake right before one) or quite abstract or conceptual, like the goal to be truthful” or “successful”, which can imply a lot of different subgoals and is relatively flexible in terms of how it is realized (Shoda et ah, 1990; Hofmann et ah, 2011). Note also that a lot of different goals can exist within one person at the same time and contradict one another. The terms intention, will, and preference, are explicitly excluded from being entailed in the notion of a “goal”, for they will be used differently in section 2.

1.2 What this Thesis is not about

The purpose of this subsection is to further clarify the scope of this thesis by explicitly excluding some topics that one might expect to be addressed in the analysis of self-control, most prominently the freedom of the will. Further limitations within the scope of this work, in example due to a limited amount of literature that could be considered, will be addressed in subsection 1.3.3.

The Self: Albeit the term “self-control” and the mentioning of the “self” in the Baumcistcr-dcfinition above (“the self’s tendency to operate on automatic pilot”) impose the question of what exactly this “self” is and how it is defined, this thesis will not be able to answer it, as there seems to be widespread dissent on what the “self” is. Additionally, it docs not seem necessary for the arguments being put forward in the following sections to make a claim on the nature of the “self”. Therefore, for the purpose of this paper, it seems sufficient to rely on a folk psychology-undcrstand of what the self is and can therefore be considered to be synonymous to the sum of one’s values, character traits, and motives. Accordingly, the role of self-control as a character trait will not be addressed extensively - even if it has been considered an important factor in shaping one’s character. It should be clear, though, that this notion of “self” docs not refer to the corresponding health magazine or a Swedish ear brand.

(Neural) Mechanisms: Due to the limited scope of the work at hand, it will not be possible to relate the different scenarios and notions of self-control adequately to the mechanisms that underlie self-control or self-control failure on a functional or neural level¹⁵. Rather, the present considerations can be thought of as a basis to analyze findings on these mechanisms in future research. The only “mechanic” measure this thesis will rely on is a general distinction between two “modes” that the mind is operating and that these modes rely on two neural systems that arc at least partly functionally separable. This contrasts “automatic” or “impulsive” action tendencies with more considerate or reflective behavior - a distinction that can already be found in ancient Greek philosophy (sec subsection 1.5 and Freud’s personality model (Freud, 1930, 1933).

Freedom of the will: The concepts surrounding self-control, like “exercising one’s will”, “overriding an automatic action tendency”, “free action”, and the like, arc inherently related to the notion of “free will” or “freedom of the will”. This topic has drawn major interest from an empirical point of view at least since the discussion of the so-called “Libet Experiments” (Libet et ah, 1983) that indicated the onset of a “readiness potential” before the subjects were consciously aware of initiating a voluntary finger movement, thus supposedly showing a prc-dctcrmincd behavior. However, it would take too much time and space to go into a discussion that adequately addresses how freedom of will should be defined and how this relates to self-control. As, on the other hand, this topic cannot be regarded as completely irrelevant to the considerations made here (as the overlapping vocabulary that is used indicates), a brief overview will be given nonetheless. According to Mark Rowlands, the problem of freedom of will can be divided into three levels: First, the “Formation of Will”, secondly, “Volition”¹⁶, and thirdly “Freedom of Action” (Rowlands, 2004). Here, “Formation of Will” corresponds roughly to the “coming about of motives (or goals)”. The central problem here is, now, whether or not my goals are determined in the way that physical causality must have lead necessarily to me having certain motives¹⁷. “Volition” then means whether or not I am able to act in accordance with my own motives or to “pursue my will” against inner resistance.¹⁸ Lastly, “Freedom of Action” refers to the question of whether or not there are physical boundaries keeping me from pursuing my will (like being tied to a chair or otherwise forced in any way (not) to act.). John Searle points out that it is rather the lack of causal sufficiency that we perceive between these levels, which is referred to as “freedom of will”: “...I do not experience the reasons - my beliefs and desires, for example - as causally sufficient to fix one decision rather than another. In common terms, I have to ’make up my mind’. Furthermore, once I have made up my mind, I still have to actually carry out the decision I have made. I do not experience the decision as causally sufficient to produce the action I have decided upon. There is, on short, a gap between the reasons for a decision and the making of the decision, and there is a second gap between the making of the decision and the actual onset of the action. These gaps have a traditional name in philosophy, they are called ’the freedom of the will’.” (Searle, 2000, p. 7)¹⁹. In the following analysis, the reader will recognize the structure of this passage in the assessment of self-control failure in section 2. However, it will not be explicitly related to the question of freedom.

1.3 Aims and Course of Action

Having reviewed the basic definitions and assumptions of this work, the purpose of this subsection is to provide the reader with an understanding of what is going to happen in the following sections. Therefore, first the aims of this thesis will be explained and subsequently, the course of action as well as potential challenges and restrictions that must be faced.

1.3.1 Aims

As stated in the introductory section, the aims of this thesis can be summarized like this:

1. Explicating the underlying conditions needed to “attest” someone a failure of self-control and the different notions of self-control that accrue from these conditions.
2. Proposing an analytical framework that makes different “self-control scenarios” (i.c., theories, experiments, and everyday-situations) comparable on the basis of these conditions.

Obviously, these goals can only be achieved in several steps, which will shortly be outlined. Here, it should be emphasized that the ultimate goal of this work is to contribute to the establishment of a unified nomenclature in self-control research, as vocabulary is, at times, used quite differently²⁰. This goal in turn can be thought in the service of merging different findings on different levels of analysis in order to gain a better insight into human behavior related to self-control, which can then serve to make the findings applicable, for instance in therapy or education²¹.

How and why the analysis of conditions that need to be met in order to “attest” self-control seemed a promising way to do so will be explained more thoroughly in the following subsection. What is left to note here arc some additional achievements that will be reached by undertaking such a task - more or less “on the run”, if one will so. First, by drawing on empirical research and analytical insight, it will be shown how to “marry” philosophical and empirical (that is to say, psychological or cognitive) research in order to gain greater insight into the related phenomena. Secondly, a structured assessment of self-control as necessary to provide an analytical framework that should prove fit to eciuip novices to the field with an understanding of the related vocabulary and concepts in order to autonomously conduct further research.

1.3.2 Course of Action

How, then shall we proceed with this admittedly ambitious undertaking? After recognizing the chal­lenges and limitations of the way self-control will be assessed here in the following subsubscetion, it will be necessary to locate this condition-based approach among other methods of self-control research and their relation (subsection 1.4). A brief outline of the further (historical) relevance of self-control and potential intersections to other disciplines 1.5 will then conclude section 1. The next section is devoted to the first aim of this thesis, namely the explication of the conditions that must be met in order to attest someone self-control failure. Starting with the condition of a two- goal conflict, the introduction of conditions and resulting scenarios will continue throughout the section up to the point where a satisfying formal description of a “genuine” delay of gratification scenario has been reached. In the process, means will be introduced that allow us to attest someone self-control failure even if that person is not willing to agree, thus describing means to override a person’s self-assessment, while trying to generally maintain first-person authority towards personal goals (section 2) This section contains the largest portion of analytic work in this thesis. Subse- qucntly, Section 3 will afterward describe the analytical framework that can be drawn from the distinctions and conditions established in the previous sections and suggest how this framework may be applied. Finally, section 4 will provisionally summarize the results of this thesis and suggest future undertakings.

1.3.3 Challenges

A vast and diverse literature as well as limited timely and spadal resources must lead to restrictions as to what can be done in a thesis like this. Therefore, the limitations concerning the literature used here and its use shall be outlined, as to clarify what can and cannot be expected of this work.

Most importantly, it is very well possible that, despite extensive research²², interesting literature has been missed, thus causing ignorance towards works that might have been valuable to the present undertaking as well as the construction of arguments that might have been previously made by others. Additionally, certain literature, although concerned with the topic at hand, was excluded from review when the most important information necessary for the present arguments had been assessed and it seemed modest to prevent an inflation of the scope of this thesis, which would have lead to a loss of analytical clarity. This includes any literature on questionnaires used to assess self-regulation as a character trait in differential psychology such as the “California Child Q-Set”²³ as well as literature on self-control models that could not be regarded as “two system models” (for an overview and categorization refer to: Hofmann et ah, 2009, 2011). The assessment of genuinely philosophical literature on the topic has been mainly focused on the authors J.R. Searle and H. Frankfurt²⁴. We will also avoid to intensively review the experimental designs and paradigms of the studies cited, as it is outside of the scope of this work to evaluate the empirical experiments on any statistical or measurement-theoretical grounds.

1.4 Methodology: It’s all about Distinctions

The construct of self-control itself relics, as should have become clear, on a variety of constructs - some of which have already been introduced (like goals). Now, as already noted, this thesis pursues to clarify the constructs involved by explicating conditions to attest self-control failure. How should such an analysis be able to shed light onto the different constructs around and even lay grounds for an analytical framework?

In order to answer this question, the following steps need to be taken. First, a distinction between the different levels of analysis in self-control research will be offered. Secondly, the relation of self­control failure and the definition of self-control will be briefly explained. Thirdly, the assessment of self-control (failure) scenarios as will be provided here and its relation to the other levels of analysis will be outlined.

The methods used to assess self-control can be divided into four broad categories or levels of analysis: The first can be denoted as the behavioral level, in which certain variables that potentially affect behavior arc manipulated throughout at least two groups of a sample in order to measure differences in the outcome of certain tasks (like the time they arc able to withstand a “temptation”) or in subject’s general behavior, like how much of a certain substance they consume. Now, these changes in behavior can be used to develop a general model of behavior by introducing constructs depicting “forces” or “systems” inside a person that shape their behavior in certain predictable or law-like ways. We shall term this level of analysis the cybernetic level·, prominent examples arc Freud’s personality model and, more recently, Kuhl’s PSI Theory (Kuhl and Kazcn, 2006; Hofmann et ah, 2011). The third level can be termed the neural level, which strives to relate certain behavior to brain processes or, more exactly neural correlates. Prominent means to do so are studies on patients with certain brain lcasons, electroencephalogram (EEG), magnctcnccphalogram (MEG), functional magnet resonance imaging (fMRI), or psychopharmacological studies. Now, all of these levels themselves can of course be divided into various levels of analysis with differing scales of “graininess” depending on the exact method of measurement. Obviously, they can also be combined - be it by looking for neural correlates of certain behavior or by deducing hypotheses from cybernetic models (or in any other way). What is most important for our purpose, however, is the fact that they all rely on some sort of meta-theoretical level which happens to be our fourth level of analysis. This again entails a lot of different things, for example a common understanding of what sort of evidence can be considered “significant” and therefore bearing some sort of meaning (e.g., a non­random relation between two variables). It entails also a common understanding of the vocabulary (terminology) that is used to depict certain classes of events or entities (like “Bereitschaftspotential” a.k.a. “readiness potential” (Shibasaki and Hallett, 2006), “prefrontal cortex,” or “conditioning”). This is necessary not only in order to “translate” findings from one level of analysis to another, but also to gain a common understanding of what has been measured (possibly using different methods) within one level of analysis. The results of this paper can be considered meta-theoretical, for it tries to provide a basis for the clarification of terminology within self-control research.

This leads us to the second step, the relation of self-control failure and the very definition of self-control. If the construct of self-control is the (dependent) variable that is to be measured in dependence of some other (independent) variable as outlined above, it needs to be defined what exactly the (behavioral) difference between “self-control” and “no (or less) self control” is²⁵. Simply put, understanding what “self-control failure” means is necessary to understand what “self-control” means. The tricky part about self-control is, however, that it is presumably a universal trait which is deployed by different people with different desires in different situations²⁶. At the same time, it needs to be assessed by potentially overruling first person authority (a person’s supposed “privileged assessment” towards one’s own perception, desires and intentions) towards their own behavior²⁷ ' and therefore requiring some clarification as to when this should be considered justified. By offering analytical conditions that define under which self-control failure might be generally attested, this thesis actually offers a mean to define self-control failure and therefore to define self-control. The resulting “nomenclature” will, however, not consist of some sort of statement that makes another claim on what self-control should mean and how it should be used, as the term “self-control” remains a construct. And it is the author’s conviction that such a construct derives its meaning from the way it is being used (e.g. Wittgenstein, 1953). Therefore, the resulting nomenclature will consist of a formal description of “scenarios” drawing strictly on the conditions explicated through the previous analysis and be thus rather a framework that can be used to describe and compare different situations (or experiments or theories) than a “classical” definition²⁸.

[...]

---

¹ The quotation marks indicate a preliminary use of the respective term that will later be amended by a more precise definition.

² Of course, also a number of frameworks for the integration of different theories have been proposed (i.e. Hofmann et ah, 2009, 2011: Hofmann and Kot.abe, 2012).

³In order to find out why it is self-control failure rather than just self-control that is being discussed here, please refer to subsection 1.4.

⁴ The WCST was first introduced by Grant and Berg (1948) based on a concept of Weigl (1927).

⁵Note that, even the “sense that the corrective adjustments originate within the person” can be preserved using a feedback-loop model of adaption as the organism’s search for information on the “position” regarding its relation to a particular goal (or subgoal) has to be an active undertaking, thus originating within the organism.

⁶ The assumption of at least two systems interacting with one another goes back to Freuds personality model including an “id” and a “superego” (Freud, 1930, 1933) and seems to be a necessary condition in order to generate conflicts within a person (i.e. Hofmann et. ah, 2009). It will therefore underlie this thesis.

⁷ For the sake of completeness, two remarks can be made: First., the additional half-sentence mentioned in the definition above, namely the capacity to “follow socially prescribed norms and rules” can be considered a special case of the notions of self-control that have just been discussed and will therefore be disregarded as redundant (e.g. , in a situation in which one follows a social rule like abstaining from stealing a neighbour’s property, the urge of wanting that particular object can be construed as short-term reward, while the desire to prevent punishment for an act of stealing (be it by law-enforcement or by the neighbour himself) represents a conflicting long-term aim: refer also to the cultural significance of self-control in subsection 1.5.). Secondly, the formulation “the goal of self-control is to...” seems to be somewhat unfortunate, because it implies teleologically an end (“the goal”) for a mean (excercise self-control), which serves actually another end (“st.reer behavior...in a desired direction”) and might induce confusion about the conceptual role of “goals” in self-control research.

⁸ For the sake of integrity, the four phases of the “Rubicon Model”, which resemble the process as described above, are given here: Predecisional Phase, Postdecisional Phase, Actional Phase, and Postactional Phase (i.e. Achtziger and Gollwitzer, 2007)

⁹ www.collinsdictionary.com/dictionary/english/Volition: assessed on December 12, 2012

¹⁰ Strictly speaking, merely “desiring” a certain state is not quite enough to qualify as “gratification” in the sense that it is instrumentalized in the experiments that will be considered here, as these entail that a certain (non-)desirable state is regarded as preferable over another, somehow competing one, while the approach of one excludes the attainment of the other. A more detailed discussion of the “Delay of Gratification’-paradigm delivering possibly relevant conditions for the validity of such a setting in terms of self-control will be given in section 2.

¹¹ As Harry Frankfurt, puts it. “Some people are naturally moved by kindness when they want to be kind, and by nastiness when they want to be nasty, without any explicit forethought, and without, any need for energetic self­control. Others are moved by nastiness when they want, to be kind and by kindness when they intend to be nasty, equally without, forethought, and without, active resistance to these violations of their higher-order desires.” (Frankfurt., 1998a)

¹² It. might, be true, however, that different, “self-control strategies” may be applied. Refer to section ?? for an elaboration.

¹³ Note that such an emphasis on first person authority rules out. means to assess self-control by simply counting behaviors a person engages in that are deemed unhealthy, immoral, or hedonistic and attesting someone self­control failure on these grounds.

¹⁴ As Baumeister and Tierney put it.: “People cannot, simply will themselves to be in love, or to feel intense joy, or to stop feeling guilty. Emotional control typically relies on various subtle tricks, such as changing how one thinks about, the problem at. hand, or distracting oneself” (Baumeister and Tierney, 2011)

¹⁵ Although it. should prove quite interesting to discuss what exactly enables people to exert sell-control, especially in cases of directly contradicting opinions, like an emphasis of attention-shifting as relevant for self-control, as suggested by Walter Mischel (Miscliel et. ah, 1972: Miscliel and Ayduk, 2011) versus an emphasis of cognitive resources as suggested by Roy Baumeister (Baumeister et al., 2007): “That marshmallow experiment caused some researchers to conclude that controlling attention is what matters, not building willpower, but we disagree. Yes, controlling attention is important. But you need willpower to comntrol attention.” Baumeister and Tierney (2011)

¹⁶ Obviously, this notion of “volition” is a different one than the umbrella term outlined above.

¹⁷ The position arguing that we are in fact, subjects to causality and that our goals are therefore pro-determined is called, accordingly “determinism”, whereas “libertarianism” refers to the position that we are, in some sense “free” in “choosing” our motives.

¹⁸ The position arguing that not the “coming about” or determination of my will is decisive for my freedom of will but rather the compliance of my motives and behavior is called “compatibilism”.

¹⁹ For a spectacular discussion of this topic between Searle and Libet. see: (Libet, 2001)

²⁰ It. is therefore not. surprising that Boekarts et ah, editors of the “Handbook of Self-Regulation”, regret the “fragmen­tation and lack of cohesion in the field of self-regulation research” and name “Developing a common theoretical framework and nomenclature of constructs” as the first general theme for future research (Boekarts et. al., 2000, p. 8)

²¹ This, of course, implies that it is in principle possible to “improve” self-control in one way or another. It seems reasonable to assume that this is in fact possible.

²² Initially, using both scientific (“PsycNet” of the American Psychological Association and “Web of Knowledge” of Thomson Reuters) and popular search engines (google Scliolar(R)), literature has been reviewed that was offered using the following keywords: Self-control, self-regulation, volition, cognitive control, inhibition, executive control, ego-depletion, delay of gratification, freedom of the will, willpower, temptation, temporal discounting, delay discounting, impulsivity, impulse control. Additionally, research has been conducted for particular keywords and authors that occurred during literature review.

²³ If however, “self-control” as a personality trait as assessed by such a questionnaire is mentioned, it will be indicated accordingly.

²⁴ Thus, even famous papers like D. Davidson’s “How is Weakness of Will Possible?” (Davidson, 1970) and others from the considerations made here will not be considered in the following analysis.

²⁵ Sell-control failure does not. necessarily need to be a binary function but can also be measured on a continuous scale.

²⁶ Therefore, “eating a lot of candy” cannot (or rather should not) simply be considered self-control failure, as not everyone thinks they need to diet and therefore restrain themselves in their candy-eating behavior.

²⁷In example, in the case of a drug addict who claims his addictive behavior is “truly his will” and therefore no self-control failure.

²⁸ Although it does in a sense define scenarios by stating conditions or, expressed in Wittgensteinian terms, “rules” that should be followed.