Jones, J. G., & Hardy, L. (1989). Stress and cognitive functioning in sport. Journal of Sports Sciences, 7, 41-63.

One of the major concerns of any athlete striving for a peak performance is controlling his/her internal state. A deciding factor in competitive performances is not the degree of skill of the athlete, but the ability to perform that skill under competitive stress. This paper considered the relationships between stress, cognitive functioning, and sports performance.

Unidimensional Models

The inverted-U theory relating arousal to performance with both high and low levels causing reduced performance, was rooted in the postulation that the reticular activating system served as a generalized arousal mechanism that responded to different types of stimuli and behavior. The origin of the inverted-U theory is often attributed to work by Yerkes and Dodson in 1908 [Yerkes, R. M. & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit formation. Journal of Comparative and Neurological Psychology, 18, 459-482]. Despite the longevity of this model it is surprising to learn that those original authors never attributed their observations to arousal levels.

To counter the invented-U theory it was shown that physiological, cognitive, and behavioral responses are specific to, rather than uniform across, different situations [Lacey, J. I. (1967). Somatic response patterning and stress: Some revisions of activation theory. In M. H. Appley & R. Trumbull (Eds.), Psychological Stress (pp. 14-42). New York, NY: Appleton-Century-Crofts].

The general application of the inverted-U theory is suspect because the theory has never been directly tested, the skill continuum of simple-to-complex is oversimplified, and there is no consideration of cognitive factors. It is purely descriptive rather than explanatory. The distinction between "simple" and "complex" skills based upon bodily movements and energy requirements produces only one classification. However, when cognitive demands in skill actions are considered a very different picture emerges. This simplistic theory is incapable of explaining the complex relationship between stress and performance.

In sports skills the active role taken by the individual must be considered, for example, considerations of perceived demands and perceived capability to meet task demands. This leads to the proposal that an individual's perception or cognitive appraisal of a situation is the crucial factor. Easterbrook [Easterbrook, J. A. (1959). The effect of emotion on the utilization and the organization of behavior. Psychological Review, 66, 183-201] had shown that increased arousal affects the selectivity of attention resulting in decreased performance when attention is inappropriate or inadequate.

A possible model relating performance to stress or autonomic arousal which considers the behavior of individuals was proposed as the "catastrophe curve" by Hardy and Fazey [Hardy, L. & Fazey, J. A. (1987, June). The inverted-U hypothesis - a catastrophe for sport psychology and a statement of a new hypothesis. Paper presented at the Annual Conference of the North American Society for the Psychology of Sport and Physical Activity. Vancouver, Canada].

  1. Under low levels of stress and physiological arousal individuals improve in performance as these two factors increase.
  2. When the performer perceives an imbalance between the task demands and his/her capability to match them, anxiety occurs and performance deteriorates rapidly and dramatically.
  3. After such failure, the original level of performance can only be regained if stress levels are considerably reduced to the point where the original performance curve was highest, that is, when the task was performed and capably handled.
  4. As performances deteriorates with increased stress/arousal there comes a point where the task is abandoned.

It assigns to cognitive anxiety the role of a splitting function that determines exactly what the effect of physiological arousal (on the day of the event) will be. When cognitive anxiety is low the model proposes that physiological arousal has only a relatively small and possibly symmetrical effect upon performance. When cognitive anxiety is high, the effect of physiological arousal is both large and catastrophic, that is, after reaching an optimal level performance is disrupted dramatically.

Multidimensional Models of Stress and Performance

Anxiety-based models of stress. These focus on the anxiety response that accompanies high levels of stress. Anxiety involves cognitive and somatic components. Cognitive anxiety (worry) is characterized by negative expectations, lack of concentration, and images of failure whilst somatic anxiety refers to physical symptoms such as nervousness and tension. This two-dimensional concept was later expanded to involve three components: cognitive, physiological, and overt behavioral. Those three components may interact such that changes in one response component may affect changes in the remaining components.

The fundamental theoretical prediction of the two- and three-dimensional models is essentially the same. The human organism has a limited channel capacity, so disruptive information in a channel is best removed by saturating the channel with neutral or relevant information. An athlete experiencing cognitive anxiety would be best served by a cognitive intervention strategy, whilst athletes experiencing unwanted physiological arousal would be best served by a somatic intervention strategy. This is the so-called "matching hypothesis."

Anxiety disrupts performance by clogging channels with useless information. Interventions should be developed to combat the symptoms of anxiety rather than its cause.

However, there is now a considerable body of experimental evidence supporting the independence of cognitive and somatic anxiety. Cognitions in athletic arenas are not solely restricted to anxiety or its absence. Potentially important meta-cognitive variables are: perceived control, self-image, and self-confidence. Bandura [Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215] proposed self-efficacy as an alternative theoretical basis to anxiety. Self-efficacy (situationally-specific self-confidence) is comprised of four elements in order of importance: i) performance accomplishments, ii) vicarious experience, iii) verbal persuasion, and iv) physiological arousal.

Jones and Hardy summarized these three orientations as follows:

"These systems correspond approximately to cognitive anxiety, self-confidence and physiological arousal. The major prediction of the cognitive-somatic anxiety literature is that interventions should focus upon whichever is the more dominant in the stress response. Conversely, the major prediction of the self-efficacy literature is that interventions must be long-term in nature and build upon positive experiences of stress that the performer has had. In particular, it suggests that structured learning situations in which the performer always succeeds, and which employ participant modeling, mental rehearsal and forms of vicarious experience, are likely to be the most powerful strategies." (p. 49)

Multidimenionsal activation states. As a rejection of unidimensional stress-performance models, it was argued that different stressful situations create qualitatively different cognitive activation states, which then influence performance via different cognitive components or processes. This is best considered not as a quantitative state but as a patterning of the physiological state. Performance efficiency is determined by the appropriateness of the patterning with respect to the task in question. This approach seems to imply that interventions should focus upon developing information-processing strategies which performers experiencing stress can use, rather than emotional-control strategies to reduce the amount of stress experienced. An example would be that through role-playing relevant "distractions," such as other contestants and bad refereeing decisions, could be gradually reduced through desensitization.

This particular model has been criticized while demonstrating some successes in particular experiments. It appears to be more relevant to verbal tasks than physical tasks.

Stress effects upon energetical and structural components of performance. Pribram and McGuinness [Pribram, K. H., & McGuinness, D. (1975). Arousal, activation and effort in the control of attention. Psychological Review, 82, 116-149] identified three separate but interacting neurophysiological systems involved in the control of attention.

  1. An "arousal" system located in the reticular formation and anterior hympothalamus and controlled by the amygdala. It controls all phasic physiological responses and in particular the orienting reflex. It is proposed as affecting or energizing perceptual processes.
  2. An "activation" system in the medial forebrain bundle, lateral hypothalamus, and basal ganglia which is responsible for tonic readiness to respond. It is proposed as affecting motor preparation processes.
  3. An "effort/coordination" system centered in the hippocampal and septal areas which coordinates the arousal and activation systems to establish relationships between perception and effort (i.e., decision-making). It is proposed as the source of demanding effort on the part of the organism.

These were integrated into a model by Sanders [Sanders, A. F. (1983). Towards a model of stress and human performance. Acta Psychologica, 53, 64-97].

"Stress will arise whenever the effort mechanism is either seriously overloaded or fails to accomplish the necessary energetical adjustments. Stress may arise because effort fails in correcting too low or too high a level of arousal, too low or too high a level of activation, or because there is a failure to supply sufficient energetical resources to decision-making. Arousal is a response to input whilst activation is viewed as being instrumental in preparing the action." (p. 54)

Many sports skills require an athlete to carry out cognitive functions as quickly and efficiently as possible. For example, effective perception of a ball in flight is affected by a player's arousal state, effective decision-making in the form of choosing an appropriate shot is influenced by effort, and cognitive preparation of the shot is affected by state of activation. Sanders' model moves from general arousal to the consideration of specific states and how they affect different components of performance. The following have been supported by research:

  1. a low level of arousal slows perceptions;
  2. a high level of arousal speeds up perception;
  3. a low level of activation slows motor preparation; and
  4. a high level of activation speeds up motor preparation.

Sanders also included an evaluation mechanism that receives feedback about the state of the system so that effort can attempt to restore any imbalance in state levels. A very high level of arousal can result in an "overflow" to the activation system so that decision-making is by-passed. This should cause faster responses, which will be beneficial in high stimulus-response compatibility situations or when the skill is well-learned. However, it also carries the danger of causing errors through inappropriate responses. This would seem to be a theoretical justification for the coaching lore that "athletes should not be asked to do new things at competitions."

Personality as a mediating variable. Some attempts have been made to relate general performance factors to personality factors and other variables. The difference between physiological stressors (e.g., time of day, sleep loss) which affect arousal and cognitive stressors (e.g., incentives, importance) which affect on-task effort is generally acknowledged. However, those factors are then moderated by a simple bi-dimensional personality theory and a dualistic classification of tasks comprised of sustained information transfer (SIT) or short-term memory (STM).

This theory [Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A theory of the relationship between individual differences and information processing. Psychological Review, 91, 1463-1474] assumes that many activities can be expressed in terms of the demands of SIT and STM resources which is unlikely given the complexity and diversity of sporting demands. However, it implies multidimensionality to understand stress and sports performance. A complex interaction exists between the psychological make-up of the individual, the nature of the stressor, and the cognitive requirements of the skill being performed.


Unidimensional models of the stress-performance relationship are too limited to be of value for understanding sporting performances. Their purely mechanical (physiological) basis is too restrictive. Without strong considerations of cognitive factors in the response, prediction will be very weak.

  1. The notion of one general arousal system is insufficient to explain stress effects upon performance.
  2. Competitive state anxiety is multidimensional comprising cognitive anxiety, somatic anxiety, and self-confidence. This requires specific interpretations of situations and the nature of an athlete's response.
  3. Stress effects are situation-specific.
  4. Stress effects are task-specific. They are dependent upon the cognitive requirements of the task so that different task components can be affected in different ways by the same stressor.
  5. Stress effects are individual- and individual-by-situation-specific. Cognitive skills and past histories of experiences will play a large role in defining the uniqueness of responses.

The relationship between stress and sports performance is an extremely complex one that involves an interaction between the nature of the stressor, the cognitive demands of the task, and psychological characteristics and abilities of the individual performing it. It is inappropriate to consider stress effects upon performance without considering the mental states brought to the scene and the cognitive activities that occur in the situation.

Return to Table of Contents for this issue.