Personality – A question of habit

In order to advance our discipline, we sometimes need to change the way the profession thinks of its central variables. We currently need to adjust our notion of personality to emphasise the routines of habit, argues Peter Warr.

William James (1890/1950) described people as ‘walking bundles of habits’. At least a third of what we do and think each day is repetition of an earlier habit (e.g. Wood et al., 2002), and most behaviours and thoughts are composites of habitual and non-habitual decision-making. Habits are widespread across single endeavours (dieting to lose weight, for instance), in social rituals, in job routines and through addictions to drug and alcohol (e.g. Clark et al., 2007).

Habits determine which situations we enter or leave, shaping what we do and think. They develop across time as people are consistently rewarded for the same activity or thought, and are strengthened by repetition. Habits free up personal resources for working through other issues, and often occur as sequences of chained thoughts and actions. Once a habit has become established, its cues in the environment or in a prior thought automatically suggest the same response. Habit-derived rewards may take the form of food, drink or social approval, but in many cases they come from an almost-instant reduction in anxiety and tension.

In addition to those behaviours and thoughts explicitly identified as habitual, two other habits important to our discipline have been disguised by their labels – attitudes and traits of personality. Attitude-based habits can be distinguished from other habits by their particularly strong emphasis on personal evaluation; they usually indicate that an individual is ‘for’ or ‘against’ something. Second, psychologists have investigated the individual who experiences a habit, drawing attention to differences between individuals in their habit-sets. In that case, we have traditionally referred to particular traits of personality.

We know from twin studies that personality traits are substantially affected by inheritance, and genetic differences between people can become consolidated through a lifetime’s repetition; personality-linked habits have occurred thousands of times. Automatic behaviours of that kind were described years ago as ‘ideomotor’.

Ideomotor habits

On 12 March 1852, Dr William Carpenter addressed the Royal Institution of Great Britain ‘on the influence of suggestion in modifying and directing muscular movement, independently of volition’. Spiritualism was very fashionable at the time, and meetings were often held to illustrate mesmerism, table-turning, water-divining and even supposedly conversing with the spirit of a deceased person. Carpenter wanted to provide a scientific explanation of those events.

In his talk, he explained paranormal behaviour in terms of an ‘ideomotor principle of action’; people respond to an imposed thought through movements which are beyond voluntary control. ‘The voluntary control over thought is entirely suspended, the individual being for the time (so to speak) a mere thinking automaton’ (Carpenter, 1852, p.147; italics in original). The label ‘ideomotor’ applies when a bodily movement immediately and automatically follows the occurrence of a linked idea.

Research in the subsequent century made it clear that, in addition to ideas prompting physical movements (through ‘ideomotor’ impact), we must also think in terms of thoughts being triggered by other thoughts; we might term those ‘ideocognitive’. I propose that psychologists would better increase understanding by viewing personality traits as composed of habits – ideomotor and ideocognitive routines triggered automatically and immediately by a single set of cues and strengthened by repetition.

Automaticity in mental activity

Carpenter’s 1852 interest was primarily in bodily actions, but automaticity also develops in mental activity. Habit researchers have often focused on overt behaviour, but we must also include thoughts and feelings. In addition to routines of behaviour, activities in the mind can also become habitual. That is the case with routines within personality; we need to supplement ideomotor responses with others which are internal, perhaps referring to those as ‘ideocognitive’. In those terms, personality traits can be viewed as bundles of ideomotor and ideocognitive habits.

Empirical support for that suggestion is necessarily in part derived from personal introspection. In Carpenter’s 1852 paper, and in many others of a similar period, cited empirical evidence was often merely an assertion of a writer’s opinion based on their introspection, for example, ‘this has been clearly proved to depend on the state of expectant attention of the performer’ (Carpenter, 1852, p.153). In respect of differences between people, Carpenter again only offered empirical support that was vague and imprecise: ‘those who have most of the power of abstraction are most easily affected in this mode; more especially if, at the same time, they are of an excitable or imaginative temperament’ (Carpenter, 1852, p.151).

After several decades when the academic psychological culture would accept as evidence only behaviours which are overt and intersubjectively confirmable, the disciplinary research pendulum is swinging back to examine mental processes. Psychologists are increasingly citing activities in the mind in their interpretation of observed behaviours.

However, procedures for gathering and citing factual evidence have massively improved since Carpenter’s talk in 1852. The quality of introspection is now very much better, with detailed investigation contrasting sharply with the casual processes of the 19th century. Today’s introspection can, for example, derive from a detailed interview about reasons for a participant’s action, or from questionnaires about feelings experienced at the time. Specific references to published evidence are now obligatory, as often are details about the reliability and validity of introspective material. Particular instances of introspection can be sub-optimal, but, as with other forms of data-gathering, a study’s introspective procedure can be either appropriate or inappropriate; psychologists’ general objection to procedures of introspection might once have been appropriate, but that is no longer the case.

Nine themes for investigating personality traits as habits

Many personality researchers have focused on the ‘Big Five’ factors of personality, often labelled as Neuroticism, Extraversion, Openness to experience, Agreeableness and Conscientiousness. (Some investigators think instead of seven primary factors of personality, splitting Extraversion into Sociability and Surgency and separating Conscientiousness into Reliability and Task Orientation.) Instead of concern about the number of factors, let us think of traits as ideomotor and ideocognitive habits, and shift our perspective in a way that may prime us to examine a different set of issues. The following nine research themes shift the view of personality as a set of traits to thinking in terms of a bundle of habits.

  • There is an extensive literature about the development and operation of habits (e.g. Aarts & Dijksterhuis, 2000; Gardner, 2015; Wood & Rünger, 2016), and we need to apply those findings and ideas specifically to the habits of personality. By setting traits within the more general habit literature, personality researchers can be supported and provoked by that wider literature. One possibility is to compare the acquisition of particular habits under different conditions of learning by people with different trait scores. Another fruitful research theme is the differential impact of particular rewards; can differences in the amount or type of reward explain variations in the development of a personality habit?
  • The habitual thoughts and feelings of high trait-scorers and low trait-scorers require detailed examination. In what specific terms do the two groups construe their current situation? We would, for instance, expect extraverted individuals to favour situations which encourage social contact and high scorers on reliability to particularly value the tidiness of a setting. But now we need more detailed understanding, including about between-person differences in the constructs applied and the emphases used. For instance, do high and low trait-scorers differ in their typical personal constructs, and do they differ in their emphasis on personal evaluation through self-referencing constructs? Other themes which might differ between high and low trait-scorers include the number and type of potential practical implications, sometimes viewed in terms of ‘affordances’; do different kinds of people envisage different situational affordances?
  • Among mental processes that become more open to study when viewing traits as habits are anticipations about likely consequences. What do different personalities expect to follow from habitual and other routines? We might predict that different personality scores are associated with increased perceived probabilities of trait-associated and other outcomes, but detailed patterns are still to be identified.
  • We need a stronger emphasis on experimental designs, conducting research to change trait-linked habits and aiming for causal explanations of observed correlations. Viewing personality as a set of habits initially requires descriptive research into the characteristics of those habits, but causal interpretations can be made through experiments to modify personality routines. Psychologists should more often seek out experimental possibilities to investigate techniques to change a habit. For instance, what situational and personal features are associated with the successful modification of habits within a personality trait?
  • What about trait development? In the 20th century, psychologists routinely investigated features associated with the growth of habit-strength, but attention was then directed to rats in mazes. We now need to examine those features in human samples. For this and other article themes, we will need to develop an acceptable index of habit-strength. And, in addition to processes of strengthening what activities are salient when someone breaks a habit?
  • In respect of particular habits, we need to examine features such as rated automaticity, flexibility or frequency of occurrence. How do individuals with high or low trait-scores perceive those and other processes? What about the impact of different environments?
  • It is also important to identify the causal triggers for particular habits. What specific cues are important to generate particular habits? Are those present in a person’s external environment or do they derive from earlier thoughts? Can influential triggers be grouped into conceptual themes (the Big Five, for instance), and are different trigger-themes associated with specific kinds of personality?
  • Shifting from traits to habits can encourage researchers to test previously developed theories about the acquisition of new habits. For example, the ‘opponent-process’ theory of motivation is central to learning and affect (e.g. Landy, 1978). The theory proposes that, although repeated exposure to a situation promotes learning, that process is automatically opposed by mechanisms which reduce the intensity of a reward. However, the automatic opponent processes are assumed to be more sluggish than direct learning, and their slower pace means that they are typically less influential than learning itself. The model has been successfully applied to drug addiction, cigarette smoking, colour vision and forms of associative learning, and is now ripe for investigation in respect of personality routines.
  • Several of these suggestions are likely to yield research papers that are largely descriptive and thus fail to meet the typical journal requirement for an investigation to be derived from and build on a particular theoretical framework. My final proposal concerns desirable research styles in novel versus established areas of investigation. In new areas of factual ignorance (as illustrated here), descriptions of current processes, for instance detailing the experiences of high versus low trait-scorers, can be of great value. However, in an established area, detailed model-testing would be more appropriate. As is frequently pointed out, one size doesn’t fit all

- Peter Warr is Emeritus Professor at the Institute of Work Psychology, the University of Sheffield.  [email protected]


Aarts, H. & Dijksterhuis, A. (2000). Habits as knowledge structures: Automaticity in goal-directed behavior. Journal of Personality and Social Psychology, 78, 53-63.

Carpenter, W.B. (1852). On the influence of suggestion in modifying and directing muscular movement, independently of volition. Proceedings of the Royal Institution of Great Britain, March 12, 147-153.

Clark, F., Sanders, K., Carlson, M. et al. (2007). Synthesis of habit theory. Occupation, Participation and Health, 27, 7S-23S.

Gardner, B. (2015). A review and analysis of the use of ‘habit’ in understanding, predicting and influencing health-related behaviour. Health Psychology Review, 9, 277-295

James, W. (1890/1950). The principles of psychology. Henry Holt. (Retrieved from

Landy, F.J. (1978). An opponent process theory of job satisfaction. Journal of Applied Psychology, 63, 533-547.

Wood, W. & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology, 67, 289-314.

Wood, W., Quinn, J.M. & Kashy, D.A. (2002). Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 83, 1281-1297.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber