Degree Name

Doctor of Philosophy


School of Management and Marketing - Faculty of Commerce


This dissertation explores the hypothesised role of psychobiological personality dimensions in shaping specific health-related behaviours and propositions potential health communications. The primary aim was to investigate Eysenck’s and Gray’s personality dimensions in relationship to specific smoking, alcohol-use, substance-use, physical exercise and eating behaviours. Past research has either focussed on a narrow range of behaviours or examined the relation between personality, health behaviours and message interventions insufficiently. It is argued that most health behaviours are a function of psychophysiological processes and social learning, and future research should find them differentially conditionable by appropriate punishment- or reward-oriented messages.

The dissertation draws on the Hippocrates-Galen-Avicennic tradition and the works of Pavlov, Skinner, Mowrer, Eysenck and Gray, and examines the previously ignored work of Avicenna on human temperament and emotions. A taxonomy of health-related behaviours and a framework for health communication and behaviour change are presented. The behavioural taxonomy of Indulgent-Denial, Delinquent-Inert, and Escape-Maintenance is assessable by physiological and psychological markers, and corresponds to Eysenck’s three-dimensional personality structure. The partly validated model explains behavioural variations by composites of nervous conduction and reactivity of behavioural systems in the brain, and also outlines possible communication interventions.

The dissertation is anchored in positivism, assuming intrinsic biogenetic influences as the psychophysiological reality. A retrospective Study consisting of a single cross-sectional survey explored the multivariate relationship between the psychobiological personality dimensions and health-related behaviours in a student sample. In line with the protocols approved by the Ethics Committee, two standard personality instruments, the Eysenck Personality Questionnaire-Revised and the Sensitivity to Punishment and Sensitivity to Reward Questionnaire, and an especially developed Behavioural Questionnaire provided the self-report measurements. Demographic effects on behavioural and personality variables were discovered by chi-square tests and two-way multivariate analysis of variance respectively, associations between variables were observed by correlational analysis, likelihood of a specific behaviour on the basis of personality variables was examined by a logistic model, and health diagnostic accuracies of personality measures were assessed by Receiver Operating Characteristic analysis.

It was found that high Extraversion and high Psychoticism scorers, and so Indulgent and Delinquent Behaviours were more likely to show appetitive associations, whereas high Neuroticism scorers and thereby Escape Behaviours were more likely to show aversive associations. All Eysenckian dimensions revealed low-to-moderate Reward-sensitivity while Neuroticism showed combined strengths of Punishment- and Reward-sensitivity. The intercorrelations of personality dimensions supported Gray that the Eysenckian dimensions probably represent unequal mixtures of Punishment- and Reward-sensitivity. The associations between personality and behaviours indicate Reward-sensitivity and Psychoticism as the most useful personality disorder dimensions, and the research questions the often-attributed role of Extraversion as a cardinal well-being dimension. Statistically and practically significant relationships between Extraversion and alcohol-use, Psychoticism and heavy alcohol-use, Psychoticism and substance-use, Psychoticism and heavy substance-use, and Neuroticism and binge-eating were observed, but the health diagnostic accuracies of personality measures were mostly poor and the correlations and odds ratios between personality and behavioural data were subject to a ceiling effect. The thesis questions whether personality measures may be a data collection method and not a diagnostic test for health-related behaviours. It is argued that personality data alone may be of insufficient diagnostic value in clinical decision-making and healthcare setting.

Thus, the dissertation asserts that health-risk behaviours will be best understood when examined in relation to the reinforcement-sensitive behavioural systems in brain and the contexts in which these behaviours occur. A psychophysiological framework of behavioural assessment and modification is advisable instead of strict personality-based models or a one-size-fits-all approach, in view of theoretical and empirical knowledge about the neuronal growth through life, nerve physiology, early childhood development, relationships between conscious and unconscious processes, the anachronistic assumption of immutability of personality traits as risk factors, results of the data analysis and the observation that health-risk behaviours are of different types and should entail differential emotional appeals. The thesis has created an architecture for future behavioural research with an emphasis on systematic punishment- or reward-oriented health communication interventions, an area that has received comparatively little empirical attention.

KEY WORDS: Behavioural Epidemiology, Behavioural Medicine, Health Communication, Personality Factors in Health and Behaviour Change, Population Health.

02Whole.pdf (3456 kB)



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.