What is Control Theory?

Concepts  from  control  theory  date  back  to  the early  1900s;  however,  the  origination  of  control  theory  is  usually  ascribed  to  the  publication  of  Norbert  Wiener’s  1948  acclaimed  work, Cybernetics:  Control  and  Communication  in  the Animal and the Machine. To describe control theory simply, one could generalize that it is premised in  understanding  and  describing  self-regulating systems.  In  the  decades  following  its  conception, control  theory  and  various  elements  of  control theory  were  used  to  guide  work  in  an  array  of diverse disciplines, including engineering, economics,  mathematics,  and  medicine.  It  wasn’t  until the early 1980s that Charles Carver and Michael Scheier  presented  control  theory  within  the  context  of  human  behavior  and  psychology,  and more specifically the regulation of health behavior through  health  psychology.  Although  introduced to  human  behavior  in  the  1980s,  control  theory has  not  been  readily  adopted  in  fields  of  health and  exercise  behavior.  As  described  by  Carver and  Scheier,  this  could  be  in  part  because  of  the acceptance and prolific use of learning theories to understand human behavior, or potentially a hesitancy  to  consider  a  theory  developed  outside  the realm  of  human  or  animal  behavior.  Regardless, the  work  of  Carver  and  Sheier  to  apply  control theory to human behavior and health psychology is the impetus that opens the door for application of control theory concepts in understanding exercise  and  physical  activity  behaviors.  The  purpose of this entry is twofold: first, to describe concepts of control theory as applied to health and specifically exercise and physical activity behaviors; and second,  to  discuss  how  these  concepts  have  been and could be used in understanding and changing exercise and physical activity behaviors.

Although most terms used in this entry will be defined at the time of introduction, there are a few that  should  be  more  specifically  discussed  at  this point. First, given the original work in control theory and subsequent application, it should be noted that control theory is also referred to as cybernetic theory or cybernetics in the literature. Clarification of  the  terms  exercise  and  physical  activity  is  also needed.  While  these  terms  are  often  used  interchangeably, each is distinct. Physical activity refers to any movement of the body produced by skeletal muscle  contraction  that  is  associated  with  energy expenditure beyond a person’s basal level. Exercise, on the other hand, is considered a subcategory of physical activity and refers to any planned, structured,  and  repetitive  physical  activity  carried  out for  the  purpose  of  improving  or  maintaining  at least one physical fitness component (cardiorespiratory  endurance;  muscular  strength,  endurance, and power; flexibility; balance; body composition; movement speed; or reaction time). For this entry the term exercise will be used, although principles of control theory are also pertinent to explaining and  changing  broader  physical  activity  behavior as well.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% OFF with 24START discount code

The Negative Feedback Loop

The foundational concept of control theory lies in what is referred to as the negative feedback loop. The feedback loop contains several important features,  beginning  with  an  input  function  (a  point where present conditions are evaluated by an individual),  also  called  perception.  Information  gathered during perception is then compared against a reference  value  (also  called  the  standard  of  comparison)  by  a  comparator.  This  reference  value represents  a  predetermined  value  of  a  condition, such as a predetermined behavioral or health goal. In this loop, the role of the comparator is similar to  the  concept  of  self-monitoring,  where  there  is an  assessment  of  current  behaviors,  any  potential external influences, and progress toward an established standard (a goal). If a discrepancy is detected during this comparison between the assessment of present conditions and the reference value, a behavior is then performed (also called output function) to  reduce  the  discrepancy  by  creating  change  in the  present  condition  so  as  to  have  an  impact  on the  environment  (anything  external  to  the  person or  system).  This  change  in  the  environment  then leads  to  a  new  perception  by  the  individual  and subsequent comparison against the established reference value. The feedback loop also includes what is referred to as disturbance, which is any change external to the individual or system (environment) that influences the feedback loop independently of the  person’s  own  action,  specifically  entering  the loop  through  environmental  impact.  This  feedback loop is a closed circuit whereby the environment,  person  (system),  and  behavior  all  influence and are influenced by one another, similar to Albert Bandura’s  construct  of  reciprocal  determinism within  the  social  cognitive  theory.  One  important note to make here is that the purpose of the feedback loop is not to create a behavior, but rather to ensure that a predetermined, desired condition (as set  by  the  reference  value)  is  maintained  through the  comparison  of  the  perception  and  reference value.  Therefore,  if  a  discrepancy  is  not  detected, no change in behavior is needed.

An example of this feedback loop can be demonstrated by considering a person who is jogging in one’s neighborhood. Once on a jog throughout the  neighborhood,  most  people  primarily  aim  to remain in a safe portion of the road, clear of cars and  other  hazards.  This  maintenance  is  usually determined and addressed through visual monitoring of the road for traffic and by establishing ample room to jog safely between regular traffic and the curb or edge of the pavement. What would happen if a large moving truck approaches the jogger or if a  shoulder  that  was  once  part  of  the  road  disappears? In using control theory with either of these scenarios,  the  jogger  would  detect  a  discrepancy between the perception of currently jogging in the road given these disturbances and the established safe jogging space reference value. For an avid jogger,  this  discrepancy  would  be  quickly  addressed by the jogger swiftly jumping off the street to continue  jogging  in  a  neighbor’s  front lawn  until  the moving  truck  passed  or  the  shoulder  reappeared. As  is  apparent  in  this  example,  during  any  given jog numerous disturbances that require behavioral actions to eliminate associated discrepancies could occur.

Reference Point/Standard of Comparison

Given  that  the  feedback  loop  heavily  relies  on  a standard of comparison, it is important to understand  the  derivation  and  nature  of  the  reference point.  As  conceptualized  by  Carver  and  Scheier, the  reference  point  is  dynamic  and  determined within  the  context  of  William  T.  Powers’s  hierarchical  system  of  goals.  In  this  hierarchy,  there are  both  higher  order  and  lower  order  goals. Achievement  of  the  lower  order  goals  is  intimately related with and essential for the achievement of the higher order goals. In this hierarchy, a higher order goal provides the reference values for feedback loops at the next lower order level. The  highest  order  level  within  this  hierarchical system  is  referred  to  as  a  system  concept,  as  it represents  both  personal  characteristics  that  one wants  to  embody  as  an  idealized  self-image  and an attempt for one to minimize sensed discrepancies  between  how  they  currently  are  versus  the ideal they desire to be. The second highest order level, principle control, specifies principles to use as  reference  in  relation  to  the  system  concept; however, these are still qualities and are not specific to any one behavior. In the next lower order level in the hierarchy, referred to as program control, actions are identified although specific steps to  accomplish  these  actions  are  not  completely defined  since  specificity  of  some  steps  depends on  situational  circumstances.  Subsequent  lower order  levels  continue  to  monitor  discrepancies between  actuality  and  goals  based  on  the  next higher  order  feedback loop. Therefore, each level within  the  hierarchy  has  independent  feedback loops that establish and adjust reference points for the  feedback  loop  of  the  next  lower  order  level, thereby  demonstrating  the  dynamic  nature  of  the goal system.

Again,  the  jogging  example  helps  illustrate this  hierarchical  goal  system.  In  this  example, the  individual  is  a  26-year-old  female,  Elizabeth, who views herself as a good friend and daughter, which  represent  the  system  concept.  Given  this view, Elizabeth usually attempts to behave in such a  manner  that  her  perception  of  her  present-self matches  this  reference  point,  demonstrating  self-regulation within this level. One example of how Elizabeth  could  accomplish  this  is  to  make  certain  that  her  mother  feels  special.  This  becomes the next lower order level reference point or goal, referred  to  as  principle  control  since  it  is  applicable  to  many  behaviors,  and  thus  could  happen in many different ways. To self-regulate with such a broad reference point, program control feedback loops are then established at the next lower order level where the principle level goal is operationalized to specific behaviors.

In this example, Elizabeth can make her mother feel special by wearing her original wedding dress for her own wedding. However, she needs to be at a lower weight so that she can wear her mother’s dress,  which  can  be  accomplished  by  exercising daily  (30–60  minutes/day).  Although  her  behavior  is  generally  defined  at  this  point  (exercising), it  also  includes  a  series  of  implicit  ifthen  decisions  based  on  the  circumstances  she  encounters. Elizabeth’s normal 45-minute jog would keep her within  the  perimeter  of  her  neighborhood;  however, if a close friend called and needed to talk she might leave the neighborhood and jog over to her friend’s house. On a different day, it could be raining, and she might choose to go up to the school gym and ride a stationary bike instead of jogging. Each of these decisions is made to ensure that the goal  of  exercising  30  to  60  minutes  daily  is  met, and each variation from the normal occurs to promote  conformity  to  any  of  her  other  goals  (e.g., not getting sick or ensuring she is a good friend). In each level of the goal hierarchy, the higher order feedback loop specifies the reference point for the next lower order level, whereby behavioral output is  required  for  successful  self-regulation  in  maintaining each reference point, and reference points are dynamic in that they reflect the feedback loop of the higher order goal.

Control Theory and Exercise Behavior

As   previously   mentioned,   the   application   of control  theory  to  exercise  behavior  is  limited, although  concepts  found  within  control  theory, especially  those  consistent  with  other  behavioral and  learning  theories,  such  as  social  cognitive theory,  self-regulation  and  the  transtheroetical model,  and  processes  of  change,  are  consistently applied to and related with exercise. Some of the most  prominent  concepts  congruent  with  control theory that have been applied to exercise are self monitoring,  specific  goal  setting,  goal  review,  the idea  of  tiered  goals  (goal  staircase),  and  performance  feedback  provision.  In  fact,  research  supports  self-regulatory  strategies  as  both  direct  and indirect predictors of exercise behavior across the lifespan.  Additionally,  interventions  that  use  self monitoring  alone  or  with  at  least  one  other  control  theory  concept  (defined  by  some  researchers as  immediate  goal  setting,  performance  feedback, and review of behavioral goals) have demonstrated a  significantly  greater  effect  on  physical  activity behavior change than those that didn’t include any of these self-regulatory strategies.


In  reading  about  how  control  theory  concepts have  been  applied  to  exercise  behavior,  it  is  evident  that  these  concepts  fall  into  the  broader theoretical context of self-regulation. Where self-regulation  is  defined  as  goal-directed  behavior, exercising control over self in terms of maintaining or achieving a preferred standard, or feedback loops,  it  is  only  successful  when  it  succeeds  at both monitoring current conditions and behavior in relation to a goal (reference point) and making changes  and  adjustments  as  desired  (based  on  a hierarchy of goals). In this context, control theory concepts are consistently supported as invaluable to understanding and initiating exercise behavior. However, given that research has heavily focused on  predictors  of  initiating  behavior,  the  roles  of self-regulation and control theory concepts might differ  when  trying  to  understand  and  improve the  maintenance  of  exercise  behavior  and  habit formation.

In  comparing  control  theory  with  many  other prominent  behavioral  theories,  two  additional points need to be made. First, where many value expectancy  and  learning  theories  postulate  that behavior change is influenced by self-reinforcement via  reward  for  goal  achievement  and  punishment  when  goals  are  not  achieved,  the  control theory does not. Rather, as described earlier, control  theory  operates  on  the  notion  that  humans inherently  self-regulate  through  the  goal-striving (reference  point  striving),  discrepancy-reduction cycle.  In  this,  reinforcement  is  only  useful  if  it provides   information   (knowledge   of   results) that  can  help  in  the  perception–goal  comparison.  Second,  the  concept  of  outcome  expectations   is   also   not   directly   addressed   here. However, this idea is widely included throughout prominent theories and works in tandem with control theory.

While  control  theory  has  much  overlap  with other  current  and  prominent  health  behavioral theories,  control  theories  have  been  met  with some  criticism  and  misunderstanding.  Some  of this  can  be  explained  through  the  historical  perspectives  described  earlier  through  the  prominent  application  of  learning  theories  and  the hesitancy to consider a theory based on a machine model.  However,  a  more  macrolevel  distinction between  control  theory  and  other  current  health behavior  theories  lies  in  the  concept  of  reinforcement.  Where  many  value-expectancy  and learning  theories  postulate  that  behavior  change is  influenced  by  self-reinforcement  via  reward for   goal   achievement   and   punishment   when goals  are  not  achieved,  the  control  theory  does not.  Rather,  as  described  earlier,  control  theory operates  on  the  notion  that  humans  inherently self-regulate   through   the   goal-striving   (reference    point    striving),    discrepancy-reduction feedback  loops.  In  this,  reinforcement  is  only useful  if  it  provides  information  (knowledge  of results)  that  can  be  integrated  and  used  within  a feedback loop.

Although  prominent  control  theory  concepts pertaining to exercise have been addressed within this entry, it should be noted that control theories include additional concepts and interested readers should see the References: list below.


  1. Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation: A control-theory approach to human behavior. New York: Springer-Verlag.
  2. Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality-social, clinical, and health psychology. Psychological Bulletin, 92(1), 111–135.
  3. Michie, S., Abraham, C., Whittington, C., McAteer, J., & Gupta, S. (2009). Effective techniques in healthy eating and physical activity interventions: A metaregression. Health Psychology, 28(6), 690–701.

See also: