Back • Return Home
MOTIVATION = LIFE
Lane Tracy
Emeritus Professor, Ohio University
106 Talon Drive, Cary, NC 27511
eMail
ABSTRACT
The study of motivation is the investigation of motives and the ways in which they cause behavior. In living systems theory a motive is defined as a strain "that energizes, activates, or moves...and that directs or channels behavior toward goals (Miller, 1978, 429)." Goal-directed behavior is one of the distinguishing characteristics of living systems. Thus, when we observe a system acting upon its environment under its own initiative, we assume (a) that it is a living sytem and (b) that its behavior is motivated. Motivation has been modeled and studied primarily at the level of organisms, especially animals. Yet the basic concepts of motivation apply to all levels of living systems. In this paper we will set forth a comprehensive model of the motivation process and examine its implications for living systems at all levels.
Keywords: living system, motive, behavior, template, decider subsystem, influence
INTRODUCTION
One of the defining characteristics of living systems is that they act to "maintain a steady state of negentropy even though entropic changes occur in them as they do everywhere else (Miller, 1978, 18)." Thus, maintenance of negentropy might be regarded as the ur-purpose of life. This original purpose then devolves into a set of more specific purposes and goals for each system, relating to the numerous variables that are important to the maintenance of the system.
A system's set of purposes and goals constitutes its values. The template of a system specifies an initial set of values that is subsequently modified and expanded through learning and maturation. The template provides a range of stability for each of the important variables, many of which are in opposition. A strain exists in the system when a variable departs significantly from its range of stability in the direction of lack or excess.
Living systems are subject to external stresses that disturb desired steady states and generate strains within the system. Strains may also be created by the system's behavior in ingesting, consuming, and extruding matter-energy and information (i.e., resource variables). Furthermore, over time the desired state for each variable must be adjusted in accordance with maturation of the system and learning of new values.
Living Systems Concept of Motives
Following Berelson and Steiner (1964), Miller (1978, 429) defined a motive as a strain "that energizes, activates, or moves...and that directs or channels behavior toward goals." In this paper motivation is defined as the process of energizing, directing, activating and maintaining behavior toward goals. (Note: Miller employs a much more restricted definition of motivation, one that does not accord with the common usage of the term.)
In some cases a motive may constitute a drive that links the strain to specific alleviatory behavior. The linkage between motive and behavior is specified in the template of the system. Some simple living systems lack behavioral options (other than death) and no choice of behavior is available or necessary. Where options do exist a drive may override other choices (e.g., the urge to duck when we perceive a flying object approaching our head). Often, however, the choice of response and even the decision on when or whether to respond is mediated by the decider subsytem.
Dominance of the decider subsystem in the motivation process emerges at the level of organisms. At lower levels, and even in simple organisms, few behavioral options may be available and template-directed behavior is the norm. In human behavior, however, choices abound and decisions must be made regarding plans of action. This continues to be true at all higher levels of living systems, i.e., groups, organizations, communities, societies, and supranational systems.
A system's values are ranged in a constantly changing hierarchy that represents "the relative urgency of reducing each of those specific strains (Miller, 1978, 34)." The hierarchy of values is employed by the decider subsystem in assessing the urgency of the various strains on the system and choosing acts that will alleviate those strains. This is a constantly ongoing process, as even temporary alleviation of all strains is a rare occurrence. The alleviation of one strain simply brings another to the fore and calls for a new choice of behavior.
Shortcomings of Current Research on Motivation
Motivation has been studied extensively at the organism level of living systems, but hardly at all at any other level. Yet it is obvious from the introductory discussion of the concept that it applies to all levels. An amoeba must alleviate strains by assessing urgency and activating goal-oriented behaviors just as a group or organization does. For that reason alone it is worthwhile to examine motivation in light of living systems theory.
Much of the research and theoretical work on motivation has been directed at isolated decisions and the behaviors activated by them. Our analysis in this paper will show, however, that this approach yields a very limited and distorted view of the process. At any moment a living system usually suffers many strains, some of which are currently being addressed by ongoing behavior. Any decision to activate new behavior must take into account the prior and ongoing choices, existing behavior, and the hierarchy of values. For instance, in deciding whether to continue reading or listening to this paper, you must consider your prior commitments to this topic (perhaps through your choice of profession), to this conference or SIG, your existing choice of how to use this time period, other strains such as hunger or thirst or tiredness, and perhaps some personal thoughts that are demanding attention. The model presented in this paper encompasses the choice of which itch to scratch as well as how hard and how long to scratch it.
Another shortcoming of current research on motivation is the tendency to isolate one system's motives from interaction with other systems. The whole point of System A's attempt to influence the behavior of B is to motivate B to do something that will relieve a strain for A. Yet research on motivation ignores influence and vice versa. The model of motivation to be presented in this paper specifically incorporates interaction between living systems.
The time dimension is seldom considered in studying motivation. Some decisions, such as marriage and career choice, have a very long time frame and tend to activate behavior long after the original choice was made. Such decisions may recognize that certain variables require periodic attention. For instance, we may buy a house with a 30-year mortgage that requires monthly payments, in the hopes that this act will take care of our need for shelter over a long period. The commitment, once made, constrains our future choices and produces new sources of strain, such as a lawn that requires weekly mowing.
Living systems are all about process and flow. Looking at motivation in terms of single motives resembles taking a bath drop by drop. We have a wide variety of needs and desires built into us by our DNA templates. These motives are augmented by learning and by our continuing association with other motivated systems. The process of life is energized and regulated by motives. To understand motivation is to understand life, but we understand very little if we examine motives one by one.
The suprasystem concept indicates that many of our motives come from higher-level systems of which we are a part, but in fact we are also influenced continually by same-level and lower-level systems. The decisions that direct our behavior are constantly being made and remade on the basis of a shifting/flowing set of motives that are derived from our own templates and from external influences. New behavior is chosen in light of past choices and current behavior. A theory of motivation that does not reflect these realities is essentially worthless. Most current theories of motivation do not even begin to recognize the complexities of the process that underlies the choice of behavior in living systems. Furthermore, the study of motivation is mired at the level of organisms despite the fact that the concept obviously applies to all levels of living systems.
A MODEL OF MOTIVATION
Prior papers have developed a living systems model of the motivation complex (Tracy 2003; 2006). That model becomes the starting point for this paper. The model will be presented briefly and then practical implications and research hypotheses will be drawn from it. The model is shown in Fig. 1.
Figure 1. Model of the Motivation Complex.
One feature of the model immediately becomes apparent. It includes many elements not usually associated with motivation, such as entropy, interactions with other systems, and reference to several of the critical subsystems of living systems. The connections among the many elements of the model are quite complicated. Ideally a model should simplify reality to focus our attention on the most essential aspects of the process. I am convinced that this model does precisely that. Simpler models have failed because too much was left out. Motivation is a very complex process and many of those complexities are essential features of the process.
How did motivation become so complex? The answer, I believe, lies in the evolutionary development of behavioral options. As organisms evolved, one route to survival was the development of new capabilities that allowed for greater adaptability. Yet new behavioral options were no advantage unless they were tied to a decision making process that made good use of them. That process involved weighing the options in the light of the strains on the system (i.e., its motives). Making an effective and efficient choice required the anticipation of outcomes and their effects on various strains, assessment of the likelihood of success, and a control process to make corrections. It also required responsiveness to the environment in order to monitor stresses, coordinate behavior with suprasystems, and assess opportunities for resource acquisition. In short, it required intelligence.
Let us focus on some of the essential complexities in this model.
1. Motivation involves interaction with other living systems. It may be possible in the laboratory to narrow the focus onto motivation processes that do not at the moment involve such interaction, but such a study is essentially barren. The whole point of a system's motivation process is to guide the behavior of the system in the direction of successful adaptation to its environment. The chosen behavior could be directed toward nonliving elements of the environment, but other living systems are a major part of the environment of any living system.
2. Motivational interaction with other living systems involves inputs and outputs of matter, energy, and information. The flow of matter and energy in and out of the system directly affects variables that are tracked in the system's hierarchy of values. Information about the flow of matter and energy is also fed back to the hierarchy of values, altering the urgency of various motives, which in turn affects decisions about the continuation of current behavior or election of new behavior. The inward flow of information may create strains that directly influence the system's motives. Outward flow of information constitutes behavior chosen to attain the goals of the system.
3. Motivation is a process, not a structure. Thus, it involves change over time. Depicting change in a two-dimensional model is not easy. You must imagine an additional time dimension through which the model is flowing and in which motives are rapidly changing. Most studies of motivation have assumed relatively constant motives, at least for the duration of the study. Yet the relative urgency of various motives is routinely altered in response to changes in the environment, interactions with other systems, the success or failure of current behavioral choices, the flow of entropic change, and the maturation or decline of the system.
4. Living systems are subject to entropy and require inputs of a variety of resources in order to maintain themselves. A system's hierarchy of values tracks the relative urgency of motives aimed at these resource variables. The variables do not disappear in this process, but may rapidly gain or lose motivational force as the urgency of motives fluctuates. Some motives tend to fluctuate rapidly, perhaps cyclically; others change very slowly.
5. Systems monitor not only the level of resources available to the system, but also the behavioral processes that are directed at regularly supplying those resources. If a process is interrupted (e.g., a person begins to choke, cutting off the flow of oxygen), the system will not wait for the level of resource to drop before choosing a new course of action. The point of this observation is that motivated behavior becomes a variable in its own right. Whether or not the system remains aware of the original purpose of the behavior, the maintenance of that behavior often becomes part of the motivational mix of variables.
6. Motives and motivated behavior are nested. Systems pursue short term goals within longer-term choices of behavior. A simple example would be a person's choice to attend a particular seminar within the choice to attend a conference, which in turn is bounded by the choice of a career or field of research. Thus, motivated behavior is highly contextual. Research on motivation often attempts to set up an experimental context within which to examine choices or the strength of commitment to a particular choice of behavior. Such research is essentially ignorant of the subject's own context and thus it fails to control an important set of variables. The model cannot directly convey the nested nature of motivational choice, but nesting relates to the time dimension and to the sequence of interactions depicted in the model.
7. The model makes it evident that much of the behavior of living systems is directed at negentropic processes. Living systems do ingest and consume raw materials, expelling waste products. These entropic acts are shown in the vertical dimension of the model. Yet much of the flow of matter, energy, and information in the model is lateral. For instance, living systems negotiate trades of resources in such a way that each system is better able to relieve its strains. From the living systems point of view they are able to make more efficient use of the resources collectively than they could have done individually. This does not impede entropy overall, but it does aid the cause of life.
8. Motivation has generally been considered to be selfish and has been studied as a self-centered process. Altruistic behavior has been treated as an anomaly. Yet the model suggests that, when two or more living systems are pursuing their own purposes and goals within a shared environment, the resulting behavior may easily be mutually beneficial. In some cases the process of evolution has encoded such behavior into the templates of the respective systems, so that we find species tending to share a niche because they naturally aid each other. In other cases living systems may arrive at the same result through trial and error (Axelrod 1984) or negotiation. Whatever the mechanism, it appears that the motivation process is sufficiently complex to allow for temporary or even semi-permanent alliances for mutal benefit. Thus, we must interpret the system boundaries in the model rather loosely. When long-term alliances form, they can easily transform into new living systems. Like almost everything else in the model, system boundaries are subject to change.
9. The level of the central system in the model is unspecified, as are the levels of all the interacting systems. As the model indicates, the interactions are not necessarily on the same level at all. Many important interactions in the motivation process are with suprasystems and subsystems. A suprasystem specifies values that its subsystems are more or less constrained to accept. The hierarchy of values of each subsystem is thus a composite, and the system's choices of behavior are directed at least in part by the motives of the suprasystem. Likewise, the motives of subsystems may influence the behavior of a suprasystem. Failure of a suprasystem to consider the well-being of its subsystems can lead to illness of the overall system. Once again, this is a matter that has hardly been touched in the literature on motivation.
10. Our focus in the model could easily shift right or left, in which case the current central system would become a source of resources and influence for another living systems. This fact highlights the interactive nature of motivation. It is like a game in which each move by one side affects the subsequent behavior of the other side. Indeed, each system may also anticipate the moves of the other and choose behavior accordingly. It should also be realized that life is seldom so simple that we can completely isolate a pair of interacting players from the behavior of other systems around them.
11. The model applies to all levels of living systems. It is likely that any hypothesis that could be constructed about human motivation, for instance, would also apply to human groups, organizations, communities, societies, and supranational systems. For example, Locke (1968) and others have tested a variety of hypothesis about the effects of goal-setting on behavioral choice and successful attainment of the goal. But all living systems from cells to societies have goals toward which they direct behavior. It might be much easier to test goal-setting hypotheses in a group or organization, where the decision-making process is more audible and visible.
12. The complexity of the interactions surely at times meets the requirements for chaos and unpredictability (Prigogine & Stengers 1984). Yet living systems generally proceed with their choices of behavior under an assumption of predictability such that a rational choice can be made. All tests of hypotheses about motivation have incorporated that assumption. A creative choice might even be rejected as being nonresponsive to the test situation. Consider that employees are often regarded as "lazy" and "unmotivated" when they do not apply a high degree of effort to their work. From the employees' viewpoint, however, the likely outcome of a decision to exert high effort may be unpredictable, whereas the payoffs from exerting low effort may seem quite predictable and acceptable. A business firm may face similar unpredictability when making a decision on whether or not to launch a new product line. Will it cause competitors to follow suit? Will customers see the new product as meeting their needs? Will they then substitute it for our other products? The variety of possible interactions with one's choice of behavior can be quite daunting.
Core Decision-making Model
The model of the motivation complex references another level of detail in the core models of decision making. One of these models, Fig. 2a, is a model of the drive process. If a system lacks behavioral options with respect to a given motive, that motive becomes a drive. When that motive becomes dominant, it leads directly to specific behavior. The behavior continues until feedback indicates that the strain is relieved.
Figure 2a. Model of template-directed motivation.
Drives are common at the level of cells, organs, and simple organisms. At higher levels, however, behavioral options become increasingly available and a more complex method of choice is required. That method, decider-directed motivation, is modeled in Fig. 2b. Fig. 2b is based on a combination of Vroom's (1964) expectancy theory and Locke's (1968) goal-setting theory, coupled with need or motive theory as represented in the hierarchy of values.
The model of the decider-directed motivation process indicates how the various interactions shown in the larger model come to influence a system's choice of behavior. The points of influence are as follows:
1. Needs (motives) are stimulated by interactions with the living and nonliving environment, and are also affected by feedback from outcomes of behavior. The amount of strain on each of the system's important variables is computed in accordance with the system's hierarchy of values (not shown), resulting in valences for expected reduction of strain being assigned to each outcome that is expected from the behavioral options being considered.
2. Valences are also subject to external influence. For instance, the system may receive persuasive communications from another system aimed at increasing or decreasing the perceived valence of certain outcomes. Promises may be made that expand an expected outcome and make it appear more valuable.
Figure 2b. Model of decider-directed motivation.
3. Valences are weighted by associations, which may consist of assumptions about the likelihood (i.e., expectancy) of attaining the outcome and/or about the usefulness (i.e., instrumentality) of the outcome toward reducing strains. But again these assumptions are open to influence from other systems, as well as from feedback of the results of prior behavior. Prior success in attaining a similar outcome increases the expectancy of success this time. Feedback from a prior outcome which, although successful, did not relieve the strain as expected would reduce the estimated instrumentality of that outcome.
4. The weighted valences of prospective outcomes from various sets of behavior (action plans) are calculated and a force toward each possible plan of action is computed. It is not intended to imply that this is entirely a conscious process or that the calculations are in any sense precise. They are based on estimates and assumptions in any case. In some instances, for instance, the system may simply update the estimates of force that were used in prior decisions. The point is that the system typically has many strains and many possible action plans to choose from, and each plan may have positive or negative effects on a variety of strains. Furthermore, the new choice may have implications for behavior that is already occurring as a result of earlier choices. It may also limit or enhance the possibility of future choices. For example, the choice to lease or purchase a location for a business may limit or broaden the possibilities for later expansion, and may also increase or decrease the strain on the current operating budget. Thus, the weighing and selection of behavioral options can be an extremely complex process, and yet it is a process that must be repeated frequently.
5. Implementation of action plans requires monitoring with respect both to the way in which the plan is carried out and to the outcomes. This is the classic control process. When the system's behavior does not match the plan or unforeseen circumstances interfere with implementation, a correction may be necessary or the plan may require modification. Such corrections are all part of motivated behavior. Furthermore, the perceptions of a plan and its implementation may be influenced by communications from other systems. Teachers, work supervisors, and counselors often seek to influence behavior through instruction or advice about the advantages of a different action plan.
6. Outcomes must be monitored and compared not only with the action plan but also with expectations of strain reduction. Both the actual outcomes and the perception of those outcomes is subject to external influence. Outcomes may be diminished or augmented by the interference of other systems and by unpredictable changes in the environment. Thus, a company that has developed a new product in expectation of large profits to offset the costs of development may find that competitors have quickly entered the market and undercut the expected outcomes. The action plan may then be modified either in the direction of cutting losses or toward increased advertising and new market development.
7. According to goal-setting theory and research, setting high goals tends to lead to better outcomes. This has been found to be true both when the goals are self-set and when they are externally influenced. Although goal-setting theory does not specify a mechanism for this effect, the model suggests that high goals tend to increase the valences of the targeted outcomes. Thus, if a teacher influences a student to aim for an A on an assignment and to choose study behavior accordingly, the strain-relieving value of the grade (such as improved self-confidence and a better chance of getting into graduate school) is enhanced.
The model of the motivation complex, in conjunction with the core models of template-directed and decider-directed motivation, presents a picture of motivation as a very complex process. As can be seen from the dates of citations above, the various pieces of the model have been around for decades. The problem is that they have not been brought together into a comprehensive picture. Instead, we have examined the elephant of motivation by separately feeling its tail, grasping its trunk, and stroking its tusks. Furthermore, observations of the elephant have not been extended to or compared with observations of other mammals or a herd of elephants.
The model may seem unsatisfactory if you are looking for easy answers to a question such as: How can I motivate my (spouse, children, students, employees...fill in the blank) to enact behavior that will improve outcomes for me or for some higher-level system? Actually, the model provides a variety of answers to that question. We have just examined a half dozen points at which decider-directed motivation is subject to influence. The problem is that, if the other system's motivation is subject to your influence, it is also subject to many other sources of influence. Furthermore, your own motivation can be similarly influenced, and indeed the very process of influencing others also helps to determine your own behavior. How can that be made simple?
Motivation is reflexive and interactive. Its effects are nested and layered and they continue over widely differing periods of time. Some behaviors are template driven while others require conscious choice. A plan of action may be repeated regularly at short intervals, long intervals, or never.
Motivation involves several different kinds of positive and negative feedback. Chaos theory suggests that, at least for higher levels of living systems, the complex feedback process may sometimes lead to instability. In fact, in the news we regularly observe evidence of motivated behavior that is difficult to understand in any other way. Even insane and criminal behavior is motivated, although the motives may seem bizarre and the behavior may seem to have no rational connection to the assumed motives. Systems are often overwhelmed by the complexity of the choices that must be made, with the result that the system may become immobilized by stress or may act irrationally.
Asking for a simple model of motivation is like asking for a simple explanation of life. A simple explanation can be concocted -- for example, we are motivated primarily by our prepotent level of needs in accordance with Maslow's (1943) need hierarchy; God created life and put humans in charge -- but it won't tell you much.
HYPOTHESES
The foregoing analysis suggests a variety of hypotheses for further study. Some of these hypotheses relate to living systems theory and the extension of the concept of motivation to all levels of living systems. Other hypotheses suggest tests of the complexities and interactions pictured by the model of the motivation complex.
1. The processes of motivation shown in the model of the motivation complex manifest themselves at all levels of living systems. This hypothesis should be relatively easy to test simply by reinterpreting existing data.
2. Decider-directed motivation is emergent at the organism level and is the dominant form of motivation at all higher levels. The control exercised by genetic templates appears to weaken with the emergence of complex organisms. For human individuals obedience to laws, moral values and cultural norms augments template-directed motivation to some extent, but the rapid expansion of behavioral options begins to overwhelm automatic responses. At higher levels formal, conscious decision processes become the norm.
3. The level of intelligence of a living system is directly related to the number of behavioral options available to that system. Some people have wondered how evolution could account for the development of human intelligence, since it seems to be more than would have been required for survival. If the development of behavioral options increased survival potential and greater intelligence was required to make good use of those options, then evolution of increased intelligence was forced upon living systems as they developed behavioral options. Demands for efficient and effective behavioral response would have favored the survival of those individuals and species that were able to make good, quick choices of action plans, and who were able to modify those choices as environmental stresses changed.
4. Motives change measurably over time. Some motives change rapidly and/or cyclically. Others change slowly and/or without any particular pattern. The tendency in motivation research has been to treat motives as fixed, at least for the duration of the experiment. For instance, when satiation occurs and a subject switches to new behavior, why do we not measure the changes in relevant motives?
5. The setting of high goals for an outcome tends to increase the valence of that outcome, thereby increasing the likelihood that an action plan directed toward that outcome is chosen. Furthermore, the chosen plan is likely to specify greater effort than would otherwise have been planned. This hypothesis is based on consistency theory. If we set a high goal, cognitive consistency requires that the expected outcome be worth the effort. Since the amount of effort is one of the variables of the plan, we tend to choose a high-effort plan that is consistent with the high valence put on the expected outcome. Tests of expectancy theory have often suffered from the assumption that the level of effort is a behavioral outcome. The model makes it obvious that level of effort is an element of the action plan, and therefore was incorporated into the calculations that led to the choice of that plan.
In addition to hypotheses the model suggests improved strategies for the study of motivation. For instance,
1. The motivation process is easier to study at the level of groups and above. In groups, organizations, communities, and societies there is likely to be open discussion of the hierarchy of values, motives, and behavioral options before a plan of action is chosen. Such discussion tends to reveal the complex interaction of motives, the assessment of valences, expectancies, and instrumentalities, the processes of external influence, and the interaction of the current decision with prior plans and ongoing behavior.
2. For purposes of control of variables it would be better to study decisions made within the envelope of a strong prior commitment, where that commitment is known, than in a situation where conflicting motives are unknown. For instance, in a study of the effects of goal-setting on work behavior it would be better to conduct the study with actual employees in a work setting than with student subjects in an artificial work context. The existing work commitments of the employees are known to some degree and can be accounted for, but the students' commitments are unknown in this situation. If the effects of commitments to current work behavior cannot be eliminated, they can at least be standardized or statistically controlled.
3. A system's existing behavior becomes a variable in its hierarchy of values. Maintenance of that behavior is an issue to be considered in the subsequent choice of other behavior. Existing action plans, including plans for long-term behavior such as marriage and career plans, must always be considered as contaminants in motivational research. Existing, and particularly long-term, action plans should themselves be an object of motivational research. The pervasive focus on change of behavior in motivational research is unwarranted. More study should be directed toward the persistence of behavior.
SUMMARY
The model of the motivation complex provides a sweeping view of the wide range of events that routinely transpire in the mundane process of moving from motives to behavior. At the level of human organisms and above, at least, our plethora of behavioral options requires complex, nuanced decision making in order to make efficient and effective choices of action plans. Our social nature, which has led to the development of several higher levels of human social systems, brings a variety of external influences to bear on the process. In particular, influence from and toward suprasystems and subsystems is a necessary part of the process. Our memories and our ability to plan ahead have added an important time dimension. All of these factors apply not only to the motivation of human individuals, but also to groups, organizations, communities, societies, and supranational systems.
The model as a whole also applies to cells, organs, and nonhuman organisms where a simpler template-directed decision process may predominate. Even at these levels interaction is required with subsystems and suprasystems. Also, lateral transfers of resources may enhance survival potential for all of the systems involved and may therefore result in establishment of collaborative behavior patterns. Lastly, the model suggests a wide variety of testable hypotheses about the motivation process, ranging from cross-level systems comparisons to connections between competing submodels of motivation. A fertile basis for further study of motivation has been provided.
REFERENCES
Axelrod, R. (1984). The Evaluation of Cooperation, Basic Books, New York.
Berelson, B., and Steiner, G. (1964). Human Behavior: An Inventory of Scientific Findings, Harcourt, New York.
Locke, E. (1968). Toward a Theory of Task Motivation and Incentives, Organizational Behavior and Human Performance, 3: 594-601.
Maslow, A. (1943). A Theory of Human Motivation, Psychological Review, 50: 370-396.
Miller, J. G. (1978). Living Systems, McGraw-Hill, New York.
Murray, H. (1938). Explorations in Personality, Oxford Press, New York.
Prigogine, I., and Stengers, I. (1984). Order out of Chaos, Bantam, New York.
Tracy, L. (2003). The Motivation Complex: Linking Needs, Decision Making, Communication, Resources, Power, and Influence, Proceedings of the 47th Annual Meeting of the ISSS, Iraklion, Greece.
Tracy, L. (2006). The Motivation Complex: A Tribute to James G. and Jessie Miller, Systems Research and Behavioral Science, 23: in press.
Vroom, V. (1964). Work and Motivation, Wiley, New York.
