Symposium organisers: Austin EJ, Saklofske DH

The emotional intelligence (EI) construct is based on the idea that people differ in the manner and extent to which they are able to experience and make use of the information available to them about their own and others' emotions. This idea is of considerable interest in the study of individual differences. It has been proposed that EI accounts for variance in a range of variables which are important to individual well-being: these include happiness, subjective life quality, success in personal relationships and career success. The possible link between EI and career outcomes has led to considerable interest in this construct in an occupational/organisational context. There are however a number of problems and unanswered questions associated with EI. These relate to the best method to measure EI (self-report or problem-based), the factor structure of EI, and the extent to which EI overlaps other constructs (e.g. certain personality traits, alexithymia, inter- and intra-personal intelligence) rather than acting as an independent predictor of the outcomes mentioned above. In this symposium both the successes of EI research and some of its more problematic aspects will be explored.

Bar-On R

Institute of Applied Intelligence, Hillerod, Denmark

A summary of findings is presented supporting the factorial structure and construct validity of the Bar-On model of emotional and social competence, which is measured by the EQ-i. The results of three factor analytic studies conducted in North America (n=3,831), Israel (n=2,670) and the Netherlands (n=1,639) are presented suggesting a 10-factor model of this construct. Findings are then presented from a number of construct validity studies cross-culturally examining these 10 factors for the degree of association with measures tapping closely related constructs. Lastly, the results of a divergent validity study conducted in North America (n=523) are presented examining the degree of association between the EQ-i and 16PF suggesting that no more than 5% of the domain measured by the EQ-i scales could be considered personality. It is, thus, concluded that the Bar-On measure is tapping what it was designed to tap which are basic factors of emotional and social competence.

Parker JDA

Department of Psychology, Trent University, Canada

Study 1 examines the empirical association between emotional intelligence (measured with the EQ-i) and the five-factor model of personality (measured using the NEO-FFI) in a large community sample of adults. Results reveal that although there are moderate associations between some specific personality dimensions (e.g., neuroticism and agreeableness), most of the variability on the total EQ-i scale is not explained by the five NEO-FFI scales. Study 2 uses the transition from high school to university as the background for examining the relationship between emotional intelligence and academic achievement. Emotional intelligence (measured with the EQ-i) was assessed during the first month of the academic year (Sept., 1999) in 417 first-year (full-time) undergraduates. At the end of the school year (May, 2000) EQ-i scores were compared between academically successful first-year students (grade-point-average greater than 80% at the end of first year) and unsuccessful first-year students (grade-point-average less than 60% at the end of first year). Although the two groups did not differ with respect to high school grade-point-average, the successful group scored significantly higher than the unsuccessful group on the total EQ-i scale

Matthews G (1), Zeidner M (2), Roberts R (3)

(1) Department of Psychology, University of Cincinnati, USA

(2) Center for the Interdisciplinary Research of Emotions, University of Haifa, Israel

(3) Department of Psychology, University of Sydney, Australia

Following a review of contemporary approaches to the assessment of emotional intelligence (EI), it would appear that self-report measures represent little more than personality, while performance-based measures do hold some promise. Nevertheless, empirical information on the latter type of measure is sparse. To redress this imbalance, a multivariate investigation, examining the psychometric properties and psychological correlates of perhaps the most promising performance-based measure of EI—the Multi-Factor Emotional Intelligence Scales (MEIS)—was conducted. Participants (N=704) completed the MEIS, the Trait-Self-Description Inventory (TSDI, a measure of the Big-Five Personality Factors), and the Armed Services Vocational Aptitude Battery (ASVAB, a measure of intelligence, widely used in the selection context). Results were equivocal. While the MEIS showed convergent validity by correlating moderately with the ASVAB and divergent validity by correlating only slightly with the TSDI, different scoring protocols currently comprising the MEIS (i.e., expert, consensus, and target) yielded contradictory (indeed opposing) findings. In addition, the reliabilities of MEIS sub-scales were often poor, with factor analyses turning up still further problems in the hypothesized hierarchical structure of the test. Based on these findings and other logical and empirical criteria, it is suggested that an emotions-more, intelligence-less approach to the study of individual differences in emotionality is warranted.

Petrides KV, Furnham A

Department of Psychology, University College London, England, UK

Two distinct types of emotional intelligence (EI) can be conceptualised based on how one attempts to measure the construct (Petrides, KV, Furnham A [2000]. On the dimensional structure of emotional intelligence. Personality and Individual Differences, 29, 313-320). Measurement by means of self-report assesses self-perceived abilities and behavioural tendencies (trait EI) whereas measurement through maximal performance instruments assesses actual ability to process emotion-laden information (information-processing EI). The present paper sets out the theoretical foundations for the operationalisation of trait EI and investigates the construct in a series of studies based on samples of adolescent pupils, university students, and adult employees. The studies provide psychometric evidence suggesting that trait EI can be isolated as a lower-order composite factor within established personality structures (Eysenckian and five-factor). In addition, wide-ranging correlational and experimental findings provide support for the construct validity of trait EI. Finally, evidence in support of the construct's incremental validity is presented by showing that, in certain cases, trait EI can account for criterion variance over and above the basic personality dimensions that partly determine it.

Saklofske DH (1), Austin EJ (2), Minski PS (1)

(1) Department of Psychology, University of Saskatchewan, Canada

(2) Department of Psychology, University of Edinburgh, UK

A short self-report emotional intelligence (EI) measure was completed by a sample of 354 students, who also completed a range of affective and personality measures. A subset of the group also completed an intelligence measure. Exploratory and confirmatory factor analysis of the EI scale suggested a hierarchical factor structure with a super-ordinate EI factor and four lower-level factors. The convergent and discriminant validity of the EI measure was investigated by examining the associations between the total EI score and the other traits measured. EI was found to be negatively and significantly correlated with Neuroticism, and positively and significantly correlated with Extraversion, Openness, Agreeableness and Conscientiousness. Correlations of EI with the other measures were in accordance with theoretical expectations, for example positive with life satisfaction and negative with depression-proneness. Most of these correlations remained significant when the effects of personality were controlled for, showing that EI accounts for variance in these measures not accounted for by personality. EI score was found not to be significantly correlated with cognitive ability.

Symposium organiser/chair: Neubauer A

Discussant: Anderson B: Linking g to the brain. Neural efficiency as a mechanism for generally better cognitive abilities

Since the 60s an increasing interest in the physiological correlates of human intelligence can be observed, motivated by the search for a biological basis of cognitive ability. The early research efforts that concentrated on speed parameters in the EEG, namely the latency of evoked potential (EP) components, have produced rather heterogenous findings. The last decade has seen a variety of new methods of EEG analysis as well as the advent of modern brain imaging techniques, like Positron Emission Tomography (PET) and EEG mapping methods, which have been also applied to the study of physiological correlates of human intelligence. The symposium features most of these new developments, e.g. the measurement of the ‘Event-Related Desynchronisation’ (ERD) and the analysis of tonic and phasic alpha power in the EEG, the calculation of entropy parameters from the EP, brain electromagnetic tomography, and the measurement of brain metabolism via Positron Emission Tomography (PET). In general, the findings of the presented studies corroborate the neural efficiency concept of intelligence, i.e. more intelligent individuals display a more efficient activation of their brains. Finally, hypotheses on eventual physical bases of the neural efficiency phenomenon shall be discussed.

Neubauer AC, Fink A, Schrausser DG

Institute of Psychology, University of Graz, Austria

In studying physiological correlates of human intelligence new brain imaging techniques like Positron Emission Tomography (PET) and EEG mapping methods focus on the level and topographical distribution of cortical activation. Actually, there is strong empirical evidence that more intelligent individuals display a more focused cortical activation during cognitive performance resulting in lower total brain activation than in less intelligent individuals. This is interpreted as a more efficient use of the cortex in high IQ individuals (‘neural efficiency hypothesis’). Former studies have used only single, homogeneous tasks and most of the studies have tested males. Therefore, here the influence of different task contents and of sex on the relationship between intelligence and cortical activation has been tested. In a sample of 26 males and 25 females we administered verbal, numerical and figural versions of a well-known elementary cognitive task, the Posner task. Generally, lower IQ participants displayed a stronger cortical activation. The figural task evoked a stronger cortical activation than the verbal and numerical tasks. Furthermore, lower IQ individuals displayed a comparatively similar pattern of cortical activation when comparing the figural, numerical and verbal test conditions, whereas higher IQ participants, primarily higher IQ females were characterized by a more distinct pattern of activation.

Klimesch W, Doppelmayr M

Department of Physiological Psychology, Institute of Psychology, University of Salzburg, Salzburg, Austria

Alpha power and cognitive performance are related to each other in a rather specific way if tonic and phasic (or event-related) alpha power are distinguished in different subbands within a frequency range of about 6 to 12 Hz. Results from three experiments indicate that (i) the extent of a phasic response in the upper alpha band (of about 10 to 12 Hz) is positively correlated with semantic memory performance, (ii) good memory performers have significantly more tonic upper alpha power than bad performers, and that (iii) tonic upper alpha power (in a resting condition) is positively correlated with intelligence (IST-70). The conclusion is that good cognitive performance is not only characterized by a large difference between tonic upper alpha power (which is typically large during rest) and phasic upper alpha power (which is typically small because alpha desynchronizes during cognitive performance) but also by a large tonic power level which in turn enhances desynchronization. We suggest that this difference, reflecting the extent of a phasic change, is related to a physiological mechanism which operates to increase the signal to noise ratio during task performance. We assume that intelligent subjects show a higher efficiency with respect to this mechanism.

Jaušovec N, Jaušovec K

Faculty of Education, University of Maribor, Slovenia

Three different methods for analyzing event-related-potentials were compared. For that purpose 74 individuals (MIQ = 107; Range 73 – 135), of average creativity performed several unattended and attended auditive and visual tasks while their EEG was recorded. In the first study the approximate entropy parameters, peak latencies and amplitudes were determined. The correlation coefficients indicated that in the attended conditions, the more intelligent individuals showed more regular ERP waveforms, decreased P300 latencies, and increased amplitudes as compared to less intelligent individuals. In the second study a comparison between high and low intelligent individuals was performed using the method of event-related desynchronization. Significant differences relating to intelligence were observed in induced and event related band power in the theta (4-7 Hz) and upper alpha band (10-13 Hz). In the third study the data were analyzed with low resolution brain electromagnetic tomography. In high intelligent individuals a decrease in the volume of activated cortical gray matter between the P300 onset and the P300 peak amplitude was observed. The EEG of low intelligent individuals showed a reverse pattern of cortical activity. The results suggest that high intelligent individuals more efficiently distributed their cognitive resources needed to cope with the tasks.

Haier RJ

College of Medicine, University of California, Irvine, USA

We conducted a PET study of emotional arousal in 22 volunteer men and women) to explore whether individual differences in g were related to the neurocircuitry of emotional experience. Each subject completed two PET scans, one while watching an emotional video and the other while watching a boring video. Each subject also completed the Raven’s Advanced Progressive Matrices (RAPM) test of non-verbal, abstract reasoning. Results showed significant male/female differences in the way RAPM scores correlated with glucose metabolic rate during the emotional and the boring conditions. In men, high RAPM correlated with basal forebrain/orbital cortex activation during the boring condition; in women, a positive correlation was found between high RAPM and activation in Wernicke’s area. In the emotional condition, men showed positive correlations between high RAPM scores and activation in hippocampal/amygdala areas, whereas the women showed positive correlations between high RAPM and activation in right hemisphere visual-spatial areas. These findings suggest that intelligence moderates the neurocircuitry activated by emotional experience differently in men and women.

Symposium organiser: Pychyl T

Chairman/discussant: Schouwenburg H

Procrastination has been investigated as an individual difference variable that has significant effects on task behaviour and performance. As a trait variable related to the ‘big five’, particularly conscientiousness, as well as aspects of the self system (e.g., self-esteem and self-regulation), procrastination is a key variable to our understanding of volitional action from an individual difference perspective. In this symposium, four papers will be presented that explore indecision, dilatory behaviour and task performance in relation to procrastination. In the first paper by Ferrari, 6 experimental studies demonstrate that decisional procrastinators process and respond to decision-making situations differently than individuals who score low on decisional procrastination measures. In the second paper, Lay considers the role of reactance in interpreting dilatory behaviour, demonstrating that trait procrastination and trait reactance predict different aspects of task behaviour which suggests that reactance may not be one of the links between trait procrastination and procrastinatory behaviour. The third paper (Pychyl) explores gender differences in the relations between procrastination, parenting style and self-worth in early adolescence. Finally, Tuckman presents an experimental study conducted in the academic context exploring the situational variables that yield an increase in the performance for students with low self-confidence.

Ferrari JR

Department of Psychology, DePaul University, Chicago, IL, USA

Effective self-regulation of performance requires a person to find an optimal balance between working on a task as fast as possible (i.e., speed) and maintaining a high degree of accuracy. Overwhelming situational demands, such as high cognitive load, task difficulty, and self-confidence in one's performance, may prompt some people to engage in self-regulation failure of effective performance speed and accuracy. It is possible in some situations persons with a tendency toward frequent indecision (decisional procrastinators) do not regulate effectively their performance speed and accuracy and, in fact, experience decisional fatigue. In the present set of experiments, indecisives compared to decisives performed simple or complex tasks, under high or low cognitive load conditions, with or without performance feedback (certainty). Across six separate experiments, compared to decisives’, results demonstrated that indecisives experienced self-regulation performance failure (low speed, low accuracy, high decisional fatigue) only in high cognitive load conditions (competing task demands), working on complex tasks, and receiving low performance feedback (uncertainty of performance). Taken together, these results demonstrate that persons who engage in decisional procrastination do not lack the cognitive ability for quality performance; they seem to lack the skill to manage effective self-regulation of their task performance.

Lay C

Department of Psychology, York University, Toronto, Canada

For one, procrastinatory behaviour may been viewed as a self-defeating attempt to establish independence. This suggests that reactance may play some role in understanding the trait and related behaviour. In the present work, however, trait reactance was related to trait procrastination in only two of four samples. In the main study, 104 university students completed a personality questionnaire and then selected one of four tasks that they preferred to complete and return the following week. Regardless of their choice, all participants were subsequently told that they must work on the Raven Matrices task. When they returned, participants indicated the extent to which they would have preferred to have worked on another task. Based solely on those who had originally not wanted to work on the Raven Matrices (N=84), trait procrastination, but not reactance, was positively associated with the number of puzzles solved and the amount of time spent on the last two days. Reactance, but not procrastination, interacted with the preference for another task in predicting overall task effort, such that increased preference for another task was negatively related to effort only for individuals high in reactance. Implications for the role of reactance in procrastination were considered further.

Pychyl TA, Coplan RJ, Reid PAM

Department of Psychology, Carleton University, Ottawa, Canada

The goal of this study was to explore the main effects and interactions between gender, maternal and paternal parenting style and global self-worth in the prediction of procrastination in adolescence. A sample of 105 adolescents (45 males, 60 females) between the ages of 13 and 15 years (Mage = 13.65, SD = .73 years) completed the Parental Authority Questionnaire Buri, 1991), the Self-Perception Profile for Adolescents (Harter, 1988) and the General Procrastination Scale (Lay, 1986). Results of hierarchical multiple regression analyses predicting procrastination revealed significant interactions between parenting styles, adolescent gender and self-worth. For females only, the effects of maternal authoritative and authoritarian parenting on procrastination were mediated through the self-system, whereas paternal parenting had a direct relation with procrastination. These gender effects in the role of the self-system are discussed in relation to parenting and procrastination.

Tuckman BW

Psychological Studies in Education Program, The Ohio State University, USA

This study focused on determining conditions to increase performance of students lacking self-confidence (self-doubters) in their ability to complete a task. Performance of these students was compared to that of students possessing self-confidence (self-believers) and those in-between (self-unsure), under competitive and individualistic conditions, on an extra credit task counting toward actual course grades. Self-confidence ratings correlated -.42 with procrastination test scores. In the competitive condition, only a fixed percentage of students could receive grade bonuses, while in the individualistic condition, any student who exceeded preset criteria could. Participants were 126 students in two demographically equivalent sections of a required psychology class, each randomly assigned to a condition. The performance task was writing test questions for the chapter assigned each week. Students were classified as self-doubters, self-believers, and self-unsure based on self-confidence ratings done twice (r=.80), prior to the start of the task. A significant interaction resulted based on self-doubters doing considerably better in the individualistic condition, self-believers performing equally in each condition, and the self-unsure doing better in the competitive condition. It was concluded that those lacking self-confidence in their ability to self-regulate are more productive in conditions where evaluation criteria are preset, thereby assuring a payoff for effort expended.

Symposium organiser/chairman/discussant: Anderson M

For 20 or so years those that have taken Spearman’s g seriously have proposed that differences in g might be best explained at the psychological level as differences in speed of information processing and at the biological level as the consequence of some global feature of cortical functioning (e.g. Jensen, 1982). Recently a contrary idea, that g is in fact is an index of the efficiency of ‘executive’ processes that are putatively supported by pre-frontal cortical areas has gained support (Duncan, 2000). At the biological level this challenges the notion that g is an index of a single global feature of neurons or neural transmission (Eysenck, 1986). At the psychological level it challenges the notion of speed as the single causal determinant of g, and generates renewed interest in ‘executive processes’ as the possible basis of Spearman’s g. Happily this also brings back a focus on mechanistic accounts of general intelligence. This symposium will examine the frontal hypothesis of g in a number of contexts including neuro-imaging and clinical studies of frontal function, general intelligence and ageing, the development of ‘executive functions’, and the link between g, emotion and the frontal lobes.

Phillips L

Department of Psychology, University of Aberdeen, Scotland, UK

Much recent literature on the frontal lobes of the brain has concentrated on the cognitive control processes¾ executive functions¾ which may be dependent on intact frontal lobe function. However, there is evidence that the frontal lobes are also involved in emotional processing, particularly the control of emotional functions. A series of studies will be described, which investigate the relationship of various emotional and executive functions with fluid and crystallised intelligence test scores. Detailed analysis of two commonly used ‘executive function’ tests: fluency, and the Tower of London task, reveal that low-g individuals do show some deficits on these tasks. However, there are substantial differences between the pattern of deficits seen in relation to g-variation in the population, and the type of deficits seen in individuals with known prefrontal injury. Studies of emotional abilities linked to the frontal lobes suggest that these may relate more highly to crystallised than fluid ability. Some frontal lobe functions do indeed appear to relate to fluid intelligence, while others relate more highly to crystallised ability this may reflect segregation of function within the large region of brain occupied by prefrontal cortex.

Anderson M

Department of Psychology, University of Western Australia, Perth, Australia

Duncan has proposed recently that cognitive processes (broadly described as executive functioning) mediated by the frontal lobes of the brain are the very processes that constitute Spearman’s g. Such a view challenges the theory that individual differences in g are based on individual differences in speed of information processing. I will argue that this contradiction can be resolved if we take seriously the idea that there are two dimensions to g, one is based on frontal functions and is related to developmental changes in ‘fluid intelligence’ and the other is based on speed of processing and is related to individual differences. Data on speed of processing and executive functioning in children and in individuals with brain damage, will be used to substantiate this claim.

Rabbitt P

Ageing and Cognitive Performance Centre, University of Manchester, UK

Attention has recently turned to theories of ‘local’ rather than ‘global’ brain ageing. We report results from two longitudinal studies of cognitive change. A sample of 5904 individuals aged from 49 to 92 years revealed that some individuals showed, over 17 years, a marked decline in memory function contrasting with stability of general intellectual ability. Implications for models of accelerated pre-frontal and temporal lobe changes are discussed. In a second study 500 individuals from the same population were repeatedly given the same battery of ‘frontal’ and ‘executive’ tests after a 5, and then after a further 3 year interval. Contrasting longitudinal changes in scores on these frontal tests and on intelligence tests, such as the Cattell and Cattell ‘Culture Fair’ tests support the ideas that, at least in some individuals, ageing brings about ‘local’ rather than ‘global’ cognitive changes, with the consequence that age-related changes in ‘executive’ functions are separable from changes in general intellectual ability (gf).

Crawford JR

Department of Psychology, King’s College, University of Aberdeen, Scotland, UK

There has been a proliferation of ‘frontal’ hypotheses in the neuropsychological literature. These hypotheses state that the behavioural changes seen in many neurological conditions (e.g., Huntington’s disease, schizophrenia, closed head injury etc) can be attributed to overarching deficits in executive functioning. However, in many cases these hypotheses have not been subject to rigorous empirical scrutiny. Often the evidence goes little beyond demonstrating that samples of patients suffering from these disorders perform significantly more poorly than controls on executive tasks. This paper will review work by the author and his colleagues that examines whether, in conditions in which a frontal hypothesis has been offered, deficits on measures of executive functioning significantly exceed deficits on measures of psychometric intelligence. That is, we test whether these deficits qualify as differential deficits. Even if deficits in executive tasks do not qualify as differential deficits, it is still possible that they make a specific contribution to impaired performance on memory tasks. We examine this by measuring the memory performance of clinical and control samples and test whether executive functioning can account for between-group variance in memory functioning even after controlling for IQ. These results are compared with those obtained when the role of the two predictors are reversed. A frontal hypothesis has also been offered to explain the cognitive changes seen in normal ageing; we have tested this hypothesis using methods similar to those outlined above.

Symposium organisers/chairs: Glicksohn J, Egan V

Zuckerman M

Department of Psychology, University of Delaware, USA

Most high sensation seekers are not psychopathic or even criminal, but all psychopaths are high sensation seekers. Sensation seeking is a primary motivation for impulsive antisocial behaviors some of which even nonpsychopathic criminals regard as irrational. Children with conduct disorders and adult psychopaths, as well as borderline personality disorders, are impulsive sensation seekers, but it takes additional traits such as aggression and asocialization to make a psychopath. A deficit in passive-avoidance learning is also related to both sensation seeking and psychopathy. Underlying the behavioral relationships are shared biological traits including: low levels of the enzyme monoamine oxidase, high levels of testosterone, and augmenting of the cortical evoked potential. The model proposes that impulsive unsocialized sensation seekers have a highly reactive dopamine system (approach), and underreactive serotonin (weak inhibition) and norepinephrine (low arousability) systems.

Weiss A (1), King JE (1), O'Connor LE (2), Berry JW (3), Enns RM (4)

(1) Department of Psychology, University of Arizona, Tucson, Arizona, USA

(2) The Wright Institute, San Francisco, California, USA

(3) Department of Psychology, Virginia Commonwealth University, Richmond, Virginia, USA

(4) Department of Animal Science, University of Arizona, Tucson, Arizona, USA

To investigate the evolutionary history of antisocial personality disorder (ASPD), this construct was assessed in a population of 154 zoo-housed chimpanzees (Pan troglodytes) using items from a 43 adjectival scale previously used to rate chimpanzee personality. The scale was constructed by clinical psychologists that read transcripts describing the behavior of chimpanzees in the sample. They then reviewed DSM-IV criteria for several major disorders that appeared evident in the population. The specific criteria were then matched to adjectives. Diagnoses requiring knowledge of a chimpanzee's cognitions were not made because these required too many assumptions. The internal consistency and interrater reliability of this scale was high. Females retained a constant low level of ASPD while males increased in ASPD over their lives. ASPD in chimpanzees was positively correlated with instances of agonistic behavior. This measure of ASPD in chimpanzees is similar to ASPD as measured in humans and shows good psychometric properties. Therefore, a behavior genetic analysis will be conducted to assess the heritability of this scale and whether any variance is due to shared zoo effects. The preliminary results show some parallels with human psychopathy and the quantitative genetic analysis will show whether, as in humans, the trait is heritable.

Blair RJR

Institute of Cognitive Neuroscience, University College London, England

I shall describe the disorder and stress its emotional component. I shall then describe the various fear positions (e.g., Patrick & Lykken) and the data presented in support of them. Empathy positions and the data in support of them will then be discussed. The fear and empathy positions both lead to clear predictions for which there is reasonably consistent support but neither position can explain the other's data. I shall suggest that it is only by reference to the anatomical level that one can resolve this conundrum. I shall consider the somatic marker theory of Damasio, concluding that this is not the core locus of dysfunction. My thesis is that psychopathy is an amygdala impairment.

Barratt E

University of Texas, USA

The rational for a conceptual neural model as a biological substrate of aggression will be considered. This will involve differentiating between two broad types of aggressive acts, impulsive or reactive and premeditated or proactive. The results of cognitive-neuroscience experiments using event related potentials (ERPs), impulsivity and anger/hostility measures as predictor or independent variables and measures of aggression as dependent variables will be presented. The conceptual neural model assumes that the cognitive/motor interface in impulsive aggression involves at least two parallel neural circuits which include the basal ganglia, amygdala, thalamus, parietal, frontal, and pre-motor cortical areas. The amygdala, basal ganglia and parietal areas are more important in impulsive aggression where impulsivity and anger play an important role. The role of these neural circuits in impulsivity per se and their involvement in impulsive aggression and antisocial behaviors will be contrasted. The ERPs were recorded during oddball and go/no go tasks and include lateralized readiness potentials. Consciousness (awareness) is intimately related to the parallel circuits and the role of the parietal lobe in differentiating the two types of aggression.

Miller E

Department of Psychology, University of Leicester, England

Brain injury is one factor that is likely to contribute to offending in some individuals. A general outline of the nature and epidemiology of closed head injury will be presented. The psychological consequences of head injury (irritability, impulsivity, lack of foresight, etc.), many of which might well make offending more likely will be stressed. Evidence will be presented indicating that frequency of head injury in offender groups is often higher than would be expected in terms of the normal prevalence of head injury. I shall discuss the very large number of methodological problems that make interpretation of the evidence difficult, including an overview of neuropsychological profiling and a comparison of the various brain-imaging modalities.

Rogers RD

Department of Psychiatry, University of Oxford, England, UK

Recent developments in cognitive neuroscience have suggested that the maladaptive and sometimes impulsive behaviour exhibited by neurological patients with lesions of the orbital frontal cortex is associated with marked deficits in the performance of laboratory decision-making tasks and may be mediated by altered processing of autonomic signals that ordinarily bias cognitive processing away from maladaptive decisions towards adaptive decisions. Our work examines the applicability of this idea to patients with severe personality disorders, and high levels of psychopathy, by investigating the relationship between impaired decision-making on a novel risk-taking procedure, multiple indices of autonomic function (e.g. electrodermal and cardiovascular) and various clinical measures (e.g. PCL-R scores). The risk-taking procedure examines behaviour in so-called ‘conflict’ situations in which participants must choose between responses that lead to small, likely rewards versus responses that lead to larger but unlikely rewards. The structure of the task permits an examination of the combination of levels of probability and levels of reward/punishment that might induce participants to take risky decisions, and its performance has already been shown to activate selectively the orbital frontal cortex using brain imaging technology in normal volunteers (Rogers RD et al. [1999]. Journal of Neuroscience, 20, 9029-9038.). The results have implications for our understanding of choice behaviour and impulse regulation in psychopathy.

Symposium organiser/chairman/discussant: Zuckerman M

Department of Psychology, University of Delaware, Newark, DE, USA

Risk taking may assume different forms: prosocial and antisocial, physical, social, legal, and financial. The antisocial and potentially self-harmful aspects of risk-taking have been overemphasised as contrasted with the prosocial, as in risky but necessary vocations (like firefighting). Then there is a type of risk-taking that is only risky for the participant and neither prosocial or antisocial, as in extreme sports. A moderate level of risk-taking is adaptive in many aspects of life. Are the same personality traits involved in all types of risk-taking or does each type demonstrate unique characteristics? The studies of Goma have addressed this question comparing fireman, criminals, and participants in risky sports. She has distinguished three personality profiles related to the different types of risk-taking. Zimmerman and Donohue have studied risk-taking in sexual behaviour and its relation to sensation seeking and impulsivity. Solomon and Ginsberg have studied risk-taking in war, comparing sensation seeking trait in Israeli soldiers decorated for bravery during the Yom Kippur war in 1973 and other veterans who had suffered from combat stress reactions.

Goma-Freixanet M

Dept of Health Psychology, Autonomous University of Barcelona, Catelonia, Spain

The idea of the existence of a continuum between personality and antisociality was generated based on the definition of the Sensation Seeking trait and the literature on personality and antisocial and prosocial behaviour. Both prosocials and antisocials engage in physical risk situations which provide varied and sometimes novel and intense sensations and experiences, and in which the participants have a risk of personal injury or death while practicing such activities. With this definition in mind we defined a continuum of physically risky activities ranging from antisociality to prosociality, the the risky sports group intermediate on that hypothetical dimension. According to this, the high sensation seekers might engage in risky situations chacteristic of any of the three types. Consequently, those persons should share certain personality features and at the same time differ on others depending on their position along the continuum. A study conducted on men identified three different profiles of risk-taking personality according to the specific traits involved in each. The same profiles were replicated in a sample of women. The three profiles were labeled: Impulsive Unsocialized Sensation Seeking, Venturesomeness, and Experience Seeking.

Zimmerman RS, Donohew L

University of Kentucky, Lexington, Kentucky, USA

In several studies with adolescents and young adults, we have investigated the independent and joint relationships of impulsive decision-making and sensation seeking with sexual risk-taking. Results have indicated strong relationships of both variables with early initiation of sexual activity, number of partners, and use of alcohol or drugs in conjunction with sex. Relationships with condom use, however, have been mixed. Sensation seeking plays a larger role in decision making in substance use with sex, and impulsive decision making plays a larger role than sensation seeking for females. We review evidence that suggests that individuals choose situations based on their sensation seeking predisiposition and make decisions once in those situations based on the extent to which they are impulsive decision-makers. Thus the combination of the two predispositions generally leads to the highest levels of sexual risk taking. We also will review the latest evidence we have collected about impulsive decision-making and the sexual decision-making process, based on an experimental study in progress. Further, we will assess the extent to which impulsivity and sensation seeking should be considered as part of one of two dimensions in predisposing individuals to risk-taking behaviour. Finally we will discuss intervention implications of these findings and results of interventions based on results to date.

Solomon Z, Ginzburg K

The Bob Shapell School of Social Work, Tel Aviv University, Israel

The current study explored the implications of sensation seeking in immediate and long term adjustment to war-related traumatic events. More specifically, the associations between sensation seeking, performance under war stress, and long term emotional adjustment, were examined. Three groups of Israeli veterans of the 1973 Yom Kippur War were studied: 112 combat stress reaction (CSR) casualties, 98 veterans who received medals for bravery, and 189 controls. Eighteen years after the war, subjects filled out a battery of questionnaires. Findings indicated that sensation seeking plays a significant role in both performance during the war and subsequent long term adjustment. Decorated war veterans were found to be higher sensation seekers than CSR casualties and controls. In addition, high sensation seekers suffered from lower levels of war-related intrusion and avoidance tendencies, and PTSD symptoms than low sensation seekers. The implications of these findings are discussed.

Symposium organisers: Netter P, Rosecrans J

This symposium will provide an overview of the biobehavioral basis of why people find tobacco products rewarding, i.e. nicotine being the suggested as the active behaviorally active chemical in tobacco. Four speakers will be present overviews of the role of nicotine in tobacco use from their own perspective, from human or basic science research obtained from experimental animal subjects. John Rosecrans will present results from experiments conducted in rats that show that the individual animal subject responds in a very individualized manner to nicotine. In general nicotine in these experiments reverse baseline-line arousal levels suggesting that nicotine may normalize behavioral extremes. David Balfour will discuss the role of brain dopamine and the potential that nicotine has to alter this behaviorally active neurotransmitter as a basis for its effects in humans. This discussion will focus on the differences between rat strains bringing into the discussion the role of genetics and the likelihood that a human subject will find nicotine rewarding. David Gilbert will further emphasize the role of differences in the individual response (i.e. Personality) of humans to tobacco and nicotine bridging the gap between basic and clinical science information presented. Petra Netter and Claudia Toll will discuss the potential role that dopamine and cortisol may play in the overall positive reinforcing effects of nicotine. The research information presented clearly shows how basic science and clinical research can be integrated and used to better understand why humans find a substance like nicotine can provide a reinforcing stimulus to humans. The overall hypothesis generated by this work suggests that, contingent upon the individual subjects makeup of their brain neurochemistry (dopamine and serotonin systems) and the subunit molecular makeup of select nicotinic receptors, an individual may find nicotine positively rewarding leading to the potential continued use of a tobacco substance.

Rosecrans JA, James JR, Venitz J

Departments of Pharmacology and Toxicology and Pharmacy and Pharmaceutics, Virginia Commonwealth

University, Richmond, VA, USA

Nicotine’s behavioral effects are most remarkable by its ability to induce rate dependent behavioral effects in both human and animal subjects. That is, this nicotinic-acetylcholinergic-receptor (nAChR) agonist, appears to be able to regulate behavior by attenuating it when baseline behavior rates are high, or by elevating behavior when baseline rates are low. Studies conducted in rodents show that these effects are especially noted in relation to differences in behavioral arousal within the same rat strain or between different rat strains. The underlying mechanism(s) that might explain these differential behavioral effects is unknown, but may be essential to understanding why select humans become dependent on nicotine via the use of tobacco. Our studies within and between strain suggest that nicotine’s variable behavioral effects are contingent on differences in the concentration, sub-type and/or function of select brain area nAChRs in these different rat populations. Furthermore, these differences may also be contingent on whether the nAChRs of a given subject susceptible to nicotine-induced acute desensitization, i.e. acute tolerance. Thus, nicotine is a unique pharmacological agent that has the ability to either activate or inactivate its receptor target depending on the state of receptor at the time of drug administration.

Balfour DJK

Department of Psychiatry, University of Dundee, Scotland, UK

It is now widely accepted that many habitual smokers find it very difficult to quit the habit because they have become dependent upon the nicotine present in the smoke. However, it is also clear that the degree of dependence experienced by individual smokers varies considerably. This presentation will focus on the evidence from an animal study that suggests that the mesolimbic dopamine system, a pathway that is thought to play a pivotal role in the development of nicotine dependence, exhibits significant strain differences in the way that it responds to repeated injections of nicotine. The data presented will show that repetitive administration of nicotine to Sprague-Dawley rats, a strain that readily self-administers the drug, causes a regionally-selective sensitisation of its effects on dopamine release in the core of the nucleus accumbens, when measured using in vivo microdialysis. In contrast the same pre-treatment protocol has no significant effects the dopamine response to nicotine in the accumbal core of Lister hooded rats. The presentation will discuss the mechanisms that may explain this strain difference and speculate on its possible psychopharmacological significance, particularly with regard to the development of nicotine dependence.

Gilbert DG

Department of Psychology, Southern Illinois University, Carbondale, USA

Evidence suggests that predispositions to smoke and to experience reinforcing effects of nicotine are partly mediated by inherited brain variations that mediate individual differences in cognition, personality, and psychopathology. Genetically influenced personality traits and variations in neurotransmitter availability and receptors modulate the effects of nicotine and quitting smoking. This presentation will focus on evidence from two sets of studies, one set assessing the acute effects of a nicotine patch versus a placebo patch on attention and the other set assessing effects of quitting smoking across 31 days. The quit data will show that smoking abstinence results in mood and EEG changes that are sustained across at least a month of abstinence and are more severe in individuals high in the neuroticism and depressive traits. The acute patch data will show that patch-administered nicotine enhances performance during a variety of attention tasks and that these enhancing effects vary as a function of task parameters and smoker traits. Abstinence and nicotine-related changes in mood, attention, and electrocortical activity will also be related to differences in CNS dopamine gene polymorphisms and models of smoking motivation.

Netter P, Toll C

University of Giessen, Germany

It is known that dopamine release is salient for reward, and that nicotine releases mesolimbic dopamine. Furthermore, effects of nicotine have been shown to be mediated by corticosteroids. Since smoking is related to approach behavior (dopamine), to impulsivity (subsensitivity of the serotonergic system), and to low susceptibility to punishment (low noradrenergic responsivity), the three transmitter systems may be related to these subcategories of smoking motivation. So the points to be touched in this contribution will be experiments demonstrating the role of dopamine for wanting and liking nicotine, the role of cortisol for the tolerance of deprivation from nicotine, and the susceptibility of the individual to the three transmitter systems dopamine, noradrenaline, and serotonin in relation to different aspects of impulsivity related to different types of smoking. Three studies were conducted in 36, 60, and 36 smokers respectively testing these effects after deprivation from smoking and being treated with transmitter-related drugs. Results show that tolerance of deprivation from nicotine is differently related to dopamine agonists and antagonists, that suppression of cortisol decreases the desire for nicotine, and that responsivity to dopaminergic, serotonergic, and noradrenergic agonists are suitable tools for identifying subtypes of smokers and of impulsivity.

Symposium organisers: Ferguson E, Patterson F

Discussant: Ackerman P

The medical profession perform an important job of public service. This symposium will examine the important role that individual differences contribute to our understanding of the selection, training and support of medical staff. The papers in this symposium will explore: (1) the interactions between learning styles, medical knowledge and career preferences as predictors of clinical training performance, (2) the role of competency based individual assessments for selecting primary care physicians, (3) the role of stress appraisals, personality and performance and (4) stress related physiological reactions and well-being. This topic is relevant to the conference as it examines the importance of a variety of individual difference measures to a wide range of outcomes including choice behaviour, appraisals, physiology, skills, knowledge and performance. In so doing a wide variety of theoretical prospective are addressed. The topic is important, not just from a theoretical stance, but also from an applied and political stance. Medical training has recently become a key political issue, with debates over equal access to medical training and medical malpractice. Therefore, the study of individual differences in medicine not only allows for theoretical issues to be explored but also shows how theory in individual differences can have important implications for policy and society.

McManus IC, Fox R, Winder BC

Research Centre for Medical Education, University College London, London, UK

Medical students differ in their approach to learning and in their study habits and learning styles. A common typology differentiates surface learning, which emphasises simple recall of factual material and is motivated by fear of failure; deep learning, with an emphasis upon understanding of principles and motivated by vocational and intellectual concerns; and strategic learning, which uses factual learning and understanding of principles selectively, resulting in patchy understanding, and is motivated by a desire for success. In this study we describe a brief 18-item version of Biggs’ Study Process Questionnaire which was used in a longitudinal study of a large sample of several thousand UK medical students who were surveyed at application to medical school in 1990 and again at graduation in 1996/7. Longitudinal factor analysis using LISREL showed long-term stability of surface, deep and strategic learning scores. Validation of the different learning styles can be seen by longitudinal analyses showing that clinical experience of patients, in the clinical years of the course, is related to study habits at application, five or six years earlier. Longitudinal analysis also shows that those students taking an intercalated BSc degree subsequently have higher deep and strategic scores and lower surface learning scores, as well as altered career preferences.

Hojat M, Veloski JJ, Gonnella JS, Erdmann JB

Center for Research in Medical Education and Health Care, Jefferson Medical College of Thomas Jefferson University, Philadelphia, Pennsylvania, USA

We examined the relationships between medical students' appraisal of stressful life events and personality profile, physical well-being, and academic performance. Total participants were 1,882 medical students who completed a set of psychosocial questionnaires. Participants reported on a 5-point scale the degree to which they had been affected by five stressful events (death or health deterioration of a family member, personal injury or health, financial and academic problems). Students who had experienced the events were divided into three groups based on their appraisal of the stressful events. Group 1 (n=446) included those who perceived that the events had influenced them a little (resilients), Group 2 (n=425) consisted of those who perceived that the events had influenced them to some extent, and Group 3 (n=420) comprised of those who perceived that the events had influenced them a lot (frails). Results of comparing the two extreme groups indicate that the so called ‘resilients’ had a more positive personality profile than the ‘frails’, reported a better physical well-being; and performed better academically. It is concluded that in the stressful environment of medical school, developing coping skills and stress management strategies can improve not only students' well-being but also their academic performance.

Patterson F (1), Ferguson E (2), Lane P (3), Norfolk T (1)

(1) Institute for Work Psychology, University of Sheffield, Sheffield, UK

(2) School of Psychology, University of Nottingham, Nottingham, UK

(3) Department of General Practice, University of Sheffield Medical School, Sheffield, UK

Research has clearly demonstrated that competency based selection systems are more effective in identifying successful job performers (e.g. Schmitt & Chan, 1998). However, this approach has rarely been applied in clinical settings (Cook, 1998). Using a multi-method design, a new competency model for general practitioners was constructed (Patterson et al, 2000). The results demonstrated that 11 competencies are critical to GP performance including empathy and communication, creative problem-solving, clinical knowledge, coping with pressure and professional integrity, amongst others. Using this model, a new selection system was developed comprising three main components; (1) a competency based application form, (2) a competency-based referencing procedure and, (3) an assessment centre comprising a collection of work-related exercises (e.g. simulation, a group exercise, a competency based structured interview), and personality and knowledge based measures. Each of these components was designed to elicit evidence for the pre-defined competencies. A predictive validation study for these new selection devices was conducted using supervisor assessments and patient ratings as the criterion indices. The evidence suggests that the new selection system was effective at identifying high performing GPs. The results are discussed in relation to medical selection more generally and future research implications are presented.

O'Connor DB

School of Psychology, University of Leeds, Leeds, UK

Occupational stress, in particular ‘high strain’ (defined as high psychological job demands and low job control; Karasek, 1979) has been found to be associated with psychological distress, coronary heart disease and hypertensive risk in a number of occupations. However, despite the large number of studies of GP stress, none have extended this model to British GPs. Two studies are described. Study one assessed job strain, levels of psychological distress and job satisfaction in a sample of GPs in the North of England (N=422), compared with a sample of white-collar workers (N=173). Study two compared twenty ‘low strain’ GPs (low demands & high control) with seven high strain GPs on ambulatory blood pressure levels during a workday and non-workday. It was hypothesised that (1) ‘high strain’ GPs would exhibit significantly greater levels of psychological distress than other GPs and the white-collar sample and, (2) ‘high strain’' GPs would exhibit heightened levels of cardiovascular arousal compared to 'low strain' GPs, with a carry-over effect into the non-workday. As predicted, ‘high strain’ GPs exhibited significantly greater levels of depressive symptoms (e.g. suicidal ideation, loss of sexual interest, feeling hopeless about the future) and job dissatisfaction than other groups, with 38% scoring equal to or above the threshold for potential clinical depression. High strain GPs’ systolic blood pressure and diastolic blood pressure was generally elevated, in particular during the non-workday, compared to their low strain counterparts. These findings are considered within the job strain paradigm and implications for patient care and future interventions are discussed.

Symposium organisers: Matthews G, Zeidner M

Discussant: Spielberger C

In 1966, Charles Spielberger's influential account of anxiety proposed that cognitive processes such as expectancies of failure, are critical to both the subjective experience of anxiety states, and their expression in behavioral measures. Since the 1960s, cognitive theories of individual differences in anxiety and other negative emotions have flourished. Contemporary theories emphasize especially the roles of appraisal and coping processes in emotion and stress, and the operation of these processes within cognitive systems for self-regulation. This symposium explores some recent developments in cognitive approaches to individual differences in anxiety and stress response. Endler presents a theoretical model that integrates some of the key cognitive constructs in anxiety research, using a multidimensional approach. He focuses especially on assessment issues and the implications of cognitive anxiety research for studies of health. Matthews and Campbell provide new perspectives on the longstanding distinction made between worry and emotionality as aspects of anxiety. They find that recent research on stress states suggests some modifications to the contrasts typically drawn between worry and emotionality. Schwean and Saklofske address an important application of work on the overlap between emotion and cognition: Attention deficit/hyperactivity disorder (AD/HD). They relate data on affect and coping in adults showing AD/HD symptoms to models highlighting difficulties in emotion-regulation as a key feature of this disorder. Ferguson focuses on the assessment-related issue of how people appraise their own coping abilities. Data from four studies demonstrate that personality biases these appraisals when they are made retrospectively. Spielberger will conclude the symposium by discussing these recent advances in theory and assessment of anxiety, coping and self-regulation.

Endler NS

Department of Psychology, York University, Canada

Anxiety, stress, and coping are key constructs for both theoretical and applied psychology and other disciplines. Towards the end of 2000, there were 63,660 citations for stress, 61,407 citations for anxiety (plus 19,538 for fear) and 128,896 citations for control, in addition to 2,215 citations for self-regulation (cf. Psyc INFO, October 12, 2000). Obviously, we are dealing with very germane constructs. The focus in this paper is on a theoretical model that integrates these various constructs and discusses the latest techniques for assessing them. Specifically, the author's multidimensional interaction model of stress, anxiety and coping will be presented, focusing on person by situation interactions and their consequences, and the effects of control on these constructs. Relevant research, including new techniques for assessing coping with health, and both separation anxiety and self-disclosure anxiety, will be presented. Various measurement techniques for the assessment of anxiety will also be presented (e.g., physiological, behavioral, ratings by others, self-report). The relevance of the model for and the effects of controllability on both physical and mental illness will be discussed.

Matthews G (1), Campbell SE (2)

(1) University of Cincinnati, USA

(2) University of Wolverhampton, UK

The distinction between worry and emotionality as aspects of anxiety is widely accepted. Typically, worry states are seen as the aspect of anxiety most detrimental to performance, whereas negative emotional states, such as tension, may be relatively benign. This paper re-examines the worry vs. emotion distinction, using the framework of the comprehensive three-factor model for subjective states developed by Matthews and colleagues. This model confirms that there are distinct factors of distress (similar to anxious emotion) and worry. Distress and worry states may also be dissociated in experimental studies of subjective response to stressors. For example, a recent study of anagram performance showed that distress is especially sensitive to task workload, whereas worry scales were also influenced by a manipulation of self-focused attention. However, our research also identifies some over-simplifications in this area. In particular, tense emotion is closely linked to cognitions of lack of confidence, so that anxious emotion should not be divorced from cognition. It is also simplistic to suppose that worry impairs performance but emotion does not; data show that relationships between both aspects of state anxiety and performance vary with task demands. We will conclude by surveying the evidence on cognitive stress processes as antecedents of worry and distress, and placing both constructs within a transactional model of stress states and self-regulation.

Schwean VL, Saklofske DH

University of Saskatchewan, Canada

Attention Deficit/Hyperactivity Disorder (AD/HD) appears to be associated with both impairments in cognitve executive functions, and difficulties in self-regulation of emotional states. Although cognitive and affective expressions of AD/HD have been mainly studied in children, it is now estimated that 40-60% of children diagnosed with AD/HD will continue to manifest the symptoms well into adulthood. This recognition has resulted in the development of several models describing the primary and secondary symptoms of AD/HD in adults. These diverse models converge on the hypothesis that poor affective modulation and stress intolerance are key features of this disorder in adults. However, research findings supporting this hypothesis are lacking. In the present study, a clinical diagnostic procedure was employed to categorize adults into one of three AD/HD subtypes, following DSM-IV, and an undifferentiated clinical group. Participants were administered a range of tests including the Millon Multiaxial Clinical Inventory, the Toronto Alexithymia Scale, and the Coping Inventory for Stressful Situations. Analyses focused on examining the relationship between primary AD/HD symptomatology, comorbidity patterns, and associated affective features in relation to coping strategies.

Ferguson E

University of Nottingham, UK

People's appraisals of the functional nature of their coping (i.e., the perceived goals associated with a particular coping strategy) are unlikely to be made dispassionately. Personality variables are likely to influence such appraisals. For example, those scoring higher on extroversion should be more likely to interpret their behaviour positively (e.g., performing an approach function). However, there is reason to hypothesise that the strength of the association between personality and perceived coping functions is time dependent. That is, the strength of this association will increase as the time lag between the actual experience of coping and its assessment also increases. Data from 4 studies (2 cross sectional and 2 longitudinal) confirm this hypothesis. Specifically, when personality and appraised coping function are assessed immediately after a stressful transaction the associations are small or zero. However, when the same assessments are made months later the size of the associations significantly increases. These effects remained once primary appraisals had been controlled and where independent of the actual type of coping behaviour adopted (i.e., the associations between personality and appraised coping function were not mediated via behaviours). These finding have implications for work on 'positive illusions' showing that personality factors can influence how the meaning of events is recalled.