1. A summary of the proposed or on-going research, its significance and results to date if appropriate.
2. A description of the purpose for which the award is required.
3. A financial breakdown of how the award is to be spent.
4. An up-to-date CV with photograph.
5. The names and addresses of two referees, familiar with the research.
6. An indication of how the applicant came to hear of the award.

 Four copies of the application, in English, should be submitted to:

In April 2008, the 9^th Annual Scholarship has been made to:

Dr. Jenny Yiend from England

On the subject of:

In April 2007, the 8^th Annual Scholarship has been made to:

Prof. Patrick Heaven from Australia

On the subject of:

Stability and change in personality: Evidence from the Wollongong Youth Study

In 2003 my colleague Joseph Ciarrochi and I commenced The Wollongong Youth Study (WYS) which involved over 700 Grade 7 students, in 5 high schools in the State of New South Wales, Australia. The main aims of the WYS are to monitor the impact of a wide range of personality and individual difference factors on their trajectories of social, emotional, and academic adjustment. Students are now in their final year of high school and, with support from the Australian Research Council, we plan to continue tracking them into adulthood. An important feature of the WYS is that we collect self- and observer reports of participants, as well as objective measures such as school grades. Thus far, observer reports have been provided by peers and teachers. Funding from the HJ Eysenck Memorial Fund in 2007 allowed us to pay for additional teacher’s reports of our participants when they were in Grade 11.

We have collected four years of teacher ratings of emotional adjustment, behavioural problems, and overall adjustment. Observations have proven to be remarkably stable across time, even though different teachers have often completed the ratings from one year to the next. Thus, ratings of emotional problems in Grade 8 correlated >.40 with ratings of the same in Grade 11. The correlations for ratings of adjustment (Grades 8 and 11) were >.30, whilst those for behavioural problems were also >.30 (all ps < .001). Personality measured in Grade 7 was found to be significantly related to teacher ratings in Grade 11. In particular, Eysenckian psychoticism (P) was found to be the best predictor of teacher ratings of adjustment and behavioural problems.

Some personality variables are measured each year and this has allowed us to monitor change and stability of personality across the entire sample. The stability-instability of personality over the course of the life span has generated considerable research interest and debate. According to Roberts and DelVecchio (2000), personality stability during the teenage years tends to be somewhat lower than during the years of emerging and early adulthood or even early childhood. It has been found, for example, that the trait consistency of the major personality dimensions is less than .50 during the teenage years, whereas it is greater than .60 for 30-39 year olds. Thus, one would expect personality scores in the WYS to exhibit similar levels of instability.

Two measures that we have repeatedly assessed in the WYS are P and conscientiousness (C). P during Grade 7 correlated .53 (p < .001) with P as assessed in Grade 10, whilst C (Grade 7) correlated .58 (p < .001) with Grade 9 C. This is evidence of some stability, but clearly stability levels are far from unity. There was also a steady decline in mean C over the course of the first three years of high school and a concomitant steady rise in mean P scores during the first four years of high school. For example, mean P scores rose from 1.78 in Grade 7 to 2.70 (Grade 10). This increase in P and decline in C is in line with other researchers who have noted declines in self-esteem over the early teenage years (e.g. Harter & Whitesell, 2003). Our findings appear to be part of a consistent pattern of findings by a number of researchers, the results of which are reflective of the various challenges associated with this period of the lifespan. We plan to continue monitoring these changes and, using growth trajectory analyses, we are now in a strong position to ascertain the extent to which changes in personality across the high school years are related to outcomes such as academic performance, observers’ ratings of behaviour, peers’ ratings of popularity, and a number of other important outcomes.

References

Harter, S. & Whitesell, N. R. (2003). Beyond the debate: Why some adolescents report stable self-worth over time and situation, whereas others report changes in self-worth. Journal of Personality, 71, 1027-1058.

Roberts, B. W & DelVecchio, W. F. (2000). The rank-order consistency of personality traits from childhood to old age: A quantitative review of longitudinal studies. Psychological Bulletin, 126, 3-25.

In April 2006, the 7^th Annual Scholarship has been made to:

On the subject of:

 Facial emotions affect the event-related potentials (ERPs) from early to late-latency components, and managing emotions are connected with personality traits. We hypothesised that the cerebral information processing of the facial expression of anger, happiness, sadness would be different among various types of personality disorders. We therefore conducted a study of these facial emotions and elicited ERPs in 38 healthy volunteers as well as in 135 outpatients with either the paranoid, schizoid, schizotypal, antisocial, borderline, histrionic, narcissistic, avoidant, dependent, or obsessive-compulsive subtype.

Their disordered personality traits were measured using the Dimensional Assessment of Personality Pathology (DAPP). Compared to the healthy control group, each patient group scored significantly different on the DAPP traits, except the schizoid group. The schizoid group scored similar to the healthy controls but significantly lower than other personality disorder groups on most of the DAPP traits. When referring to the topology of a given facial stimulus eliciting ERPs over the three midline electrode sites, the healthy control group displayed longer N1 (N170) and P2 components at Fz than those at Cz and Pz, the N2 amplitude was maximal at the Cz, the P3a and P3b amplitudes were maximal at Pz. Similar topological patterns were found for personality disorder groups.

When referring to the facial emotion effect on the ERPs in each group, neutral faces elicited longer P2 latencies in the schizoid, schizotypal, antisocial, borderline and obsessive-compulsive groups; longer N2 and P3b latencies in the borderline group, but shorter N2 latencies in the obsessive-compulsive group; and lower P2 in the obsessive-compulsive group. Angry faces elicited significantly shorter P2, P3a and P3b latencies in the healthy control group; shorter N2, P3a and P3b latencies in the paranoid group; shorter P3a latencies in the schizoid, schizotypal, and borderline groups; shorter N2 latencies in the histrionic group, but longer N2 latencies in the avoidant group; longer P3b latencies in the narcissistic and the dependent groups; higher N1 (N170) amplitudes in the avoidant group; lower P2 amplitudes in the histrionic group; and higher P3b amplitudes in the antisocial group. Happy faces elicited longer P3a latencies in the antisocial group. Sad faces on the other hand, elicited longer P2 latencies in the borderline and histrionic groups; longer N2 and P3b latencies in the borderline group.

When referring to the facial emotion effect on the ERPs in all subjects, neutral faces elicited higher N2 amplitudes in the schizotypal and the dependent groups than those in the healthy control group. Angry faces elicited shorter P3b latencies in the borderline group than those in the healthy control group. Happy faces elicited longer P3b latencies in the schizotypal group than those in the healthy control group. Sad faces elicited shorter P3a in the narcissistic group, but longer P3b in the schizotypal group than those in the healthy controls. Moreover, the healthy subjects used longer reaction times to elicit anger than other emotions, and the narcissistic patients used less reaction time to elicit sadness than other emotions. In addition, some ERP component latencies and amplitudes were correlated with DAPP traits. Our findings have shown that facial emotions (especially anger) affect both early and late ERPs, and the effects were different as to the different personality disorder types.

In April 2005, the 6^th Annual Scholarship has been made to:

On the subject of:

To date, there have been numerous genome wide scans for Neuroticism, Neuroticism-like traits and genetically related measures of mood and anxiety. Despite the interest in Neuroticism, there are no known genome wide scans for adult Extraversion or Psychoticism. Much less is known about the genetics of adolescent personality. The aim of this paper is to report a genome-wide scan based on measures of adolescent personality from Australian twins and their siblings.

Data for this study was collected as part of ongoing studies into the development of melanocytic naevi and cognition at ages 12, 14 and 16. At each interview, twins, cotwins and their siblings were asked to complete the full 81 item Junior Eysenck Personality Questionnaire (JEPQ) which assesses the three major dimensions of personality: Psychoticism, Extraversion and Neuroticism. In addition, the questionnaire contained the Lie scale which is a measure of social desirability. Complete phenotypic data were available from 2086 twin individuals and 511 siblings.

Blood or three buccal cell samples were obtained from twins, siblings and most parents for blood grouping and DNA extraction. There were 439 families with a minimum of 1 sib-pair and which had complete genotypic and phenotypic data.

Multipoint identity by descent probabilities were calculated at each of the autosomal markers using MERLIN. For each wave of data collection, variance components linkage analyses were run in Mx and LOD scores plotted. The Mx variance components method has the advantage of being able to model the effects of age and sex on the means for each of the personality dimension.

We found only one area of significant linkage on chromosome 4 for Lie. We found no linkage for Psychoticism, Extraversion or Neuroticism. There were just two regions of suggestive linkage; Extraversion on chromosome 3 and Psychoticism on chromosome 7.

An advantage of this study was the availability of multiple measurements on the same individuals across time. Normally this would provide the powerful option of performing multivariate QTL analysis to increase the power to detect QTLs which affect measurements pleiotropically. Multivariate QTL analysis can also provide supporting evidence for linkage if a signal is present in the same region at other ages. Yet despite evidence of significant additive genetic continuity across the three waves based on genetic simplex modeling using the same data, there was little congruency between the LOD score curves at 12, 14 and 16 years.

To our knowledge, this study is the first to show a genome wide scan of adolescent personality measures, and certainly the first genome wide scan for the dimensions of Psychoticism, Extraversion and Lie. We are currently working on incorporating mood and affective measures while increasing the sample size to increase power to detect linkage.

In April 2004, the 5^th Annual Scholarship has been made to:

On the subject of:

The broad aim of this study is to investigate the genetic and environmental factor structure underlying the relationship between personality traits and cognition. For instance, previous studies have shown a relationship between the NEO Personality Inventory (-Revised) 'Openness to Experience' domain and intelligence test scores, and this was explored in the present study. By using a classical twin design, the proportion of genetic and environmental variance contributing to the correlation between Openness and Intelligence could be determined.

Measures of Openness (NEO PI-R), verbal and performance IQ (Multidimensional Aptitude Battery) and academic achievement (Queensland Core Skills Test) were obtained from an adolescent twin sample (N = 274 pairs). Openness was correlated 0.26 with verbal IQ, 0.14 with performance IQ, and 0.31 with academic achievement. Academic achievement was more strongly correlated with verbal (0.81) than performance (0.55) IQ, and verbal IQ and performance IQ correlated 0.50.

Structural equation modelling of the covariation between Openness, verbal IQ, performance IQ, and academic achievement in identical and non-identical twins enabled estimation of genetic and environmental sources of covariation between measures. Based on previous findings, a genetic factor structure was hypothesised that included 1) a general genetic factor, 2) a genetic factor influencing IQ and academic achievement, 3) a genetic factor influencing verbal IQ, 4) a genetic factor influencing performance IQ, and 5) a genetic factor influencing Openness (academic achievement, in this sample, has been previously shown to have no independent genetic influences outside of those shared with IQ). A model in which the general genetic factor loading on performance IQ was dropped did not fit the data, although the general factor's influence on performance IQ (factor loading of 0.47) was smaller than its influence on verbal IQ (0.62) and academic achievement (0.80). The general genetic factor explained more covariance between the IQ and academic achievement measures than did the genetic factor specified to influence only IQ and academic achievement. This suggests that Openness is related to general ability rather than to any specific aspect of measured intelligence. As a unique genetic influence on Openness was supported (explaining 28% of variance) this indicated some separation of Openness and Intelligence as unique constructs. The results further showed that the environment (common and nonshared) did not mediate any of the covariation between Openness and IQ or academic achievement.

This analysis will be expanded to include the facets of Openness, which may show different magnitudes of genetic correlation with the IQ and academic achievement measures. Future research will also be directed to the genetic and environmental relationship between other personality domains (especially Extraversion and Conscientiousness) and cognitive abilities.

In April 2003, the 4^th Annual Scholarship has been made to:

On the subject of:

The relationships amongst emotional intelligence (EI) assessed by questionnaire (trait EI), psychometric intelligence, performance on tasks in which emotional information is processed and performance on non-emotional tasks were studied. In Study 1, 95 participants completed two trait EI measures and also performed several inspection time (IT) tasks involving the processing of both emotional and non-emotional information. An unspeeded emotion-recognition task and an assessment of fluid ability (Raven's matrices) were also included. In Study 2, 50 participants completed an emotion-recognition IT task, a trait EI measure, a verbal ability test (Mill Hill Vocabulary Scale) and the Levels of Emotional Awareness Scale (LEAS), which requires the verbal description of the emotions of oneself and another person in a number of situations.

In Study 1, factor analysis showed that the associations amongst emotion task performance, non-emotion task performance and Raven's score can be accounted for by two correlated factors: a 'speed' factor with high loadings from non-emotional IT task performance scores and Raven's score and an 'emotion' factor with high loadings from emotion task performance scores. Similar results were obtained using structural equation modelling. A composite measure of emotion task performance was found to be significantly positively correlated with an EI subscale assessing the ability to appraise emotions. In Study 2, LEAS score and the emotional IT task score were significantly positively correlated. LEAS score was also significantly positively correlated with an EI appraisal of emotions subscale; this correlation remained significant when verbal ability was controlled for. Performance on the IT task was negatively correlated with an intrapersonal EI scale assessing mood regulation.

The results from Study 1 suggest that performance on emotion-related tasks can be linked to the information-processing approach to psychometric intelligence, with the finding being that there are separate but correlated factors relating to processing speed and emotion task performance. The two studies provided some evidence for associations between trait EI and emotion task performance, an issue which is important in the validation of EI by questionnaire. The effect directions suggest that interpersonal EI (related to emotion perception) is positively associated with task performance, but intrapersonal EI (mood regulation) is negatively associated.

Further work is required to obtain information about the associations of a wider range of performance EI measures with non-emotion task performance and psychometric intelligence in order to establish more clearly how emotional abilities link to intelligence and to general processing-speed. Further studies are also required to establish the patterning of correlations between trait EI and a wider range of performance EI measures in order to obtain more information about this aspect of the validation of trait EI.

In April 2002, the 3^rd Annual Scholarship has been made to:

On the subject of:

In two studies, correlates of risky driving among young Turkish drivers were investigated. The aim of the first study was to investigate how gender (male and female) and sex-role (masculine, feminine, androgyny and undifferentiated) are associated with risky driving behaviour, traffic offences and accident involvement among young drivers. Three hundred and fifty four young drivers (221 males and 133 females) filled in a form including a short form of the BEM Sex-Role Inventory (BSRI), the Driver Behaviour Questionnaire (DBQ) and questions about a driver's accident history and background information. Participants were classified into one of the sex-role groups according to their BSRI scores by using a median-split separately for men and women. ANOVA results showed male drivers reported a higher number of aggressive and ordinary violations than the female drivers did. High masculinity score was related to high number of ordinary (highway code) violations and penalties. Among young male drivers, participants having high masculinity scores reported more aggressive and ordinary violations than other sex-role groups did. Sex-role did not have effects on driving measures among female drivers.

In the second study, the objective was to study the relationships between young drivers' life-style, personality (sensation seeking tendency) and driving behaviour. A survey containing a life-style inventory (LSI), Zuckerman's Sensation Seeking Scale (SSS), DBQ and questions about driving history (mileage, active and passive accidents, punishments), was administered to 344 participants (230 of them having a driving licence). Factor analysis was conducted on the LSI to identify the predominant life-style types. Five orthogonal life-style factors were found: religion (e.g. fasting, going to mosque), social orientation (e.g. meeting with friends in cafes, cinema), "parties and sport" (e.g. matches, sport, discos), culture (e.g. poems, Turkish folk music), and home orientation (e.g. staying at home, TV). Poisson and negative binomial regression analysis showed that none of the personality or life-style variables predicted involvement in active accidents (hit a vehicle or an object). Whereas the number of passive accidents (hit by another vehicle) was predicted positively by "culture" and "parties and sport" life-style orientations and negatively by "home" orientation. Multiple regression analysis was conducted for DBQ errors and violations scores. Number of self-reported driving errors was related positively to BS and negatively to "social" life-style orientation. DBQ violations were predicted positively by BS and Disinhibition from the SSS, "parties and sport" and religion as life-style orientation.

These two studies helped us to identify some new factors related to risky driving. Especially masculine sex-role and life-style characterised by going to discos, interest in sports and motor vehicles seemed to be related to risky driving. These results might partly explain why young male drivers are four times more often involved in fatal car accidents than the same aged female drivers.

In April 2001, the 2^nd Annual Scholarship has been made to:

On the subject of:

The objective of this study was to investigate the relationship between personality characteristics, self-reported driver behaviour (driver errors and violations) and road traffic accident involvement. It was supposed that personality factors would mainly be related to accident involvement via driving violations, i.e. deliberate deviation from safe driving. Three data sets were collected.

The first data set consisted of survey answers of 273 Turkish professional drivers who completed questionnaires including the Inventory of Sensation Seeking (ISS; Arnett, 1994), the Brief Symptom Inventory (Degoratis, 1992), the Driver Behaviour Questionnaire (DBQ; Reason et al., 1990), and questions about their accident history and driving experience. The second data set was collected among 230 young drivers (university students). The questionnaires included the DBQ, the Sensation Seeking Scale (SSS; Zuckerman, 1979), and background information (active and passive accidents, driving experience, etc.). In the third study, 172 university students completed forms including the DBQ, the ISS, the Eysenck Personality Questionnaire (EPQ; Eysenck & Eysenck, 1991/1996) and questions about their accident involvement (active and passive accidents) and driving experience.

Reliability analysis indicated that both the sensation seeking scales (SSS and ISS) had low internal consistency whereas the other scale scores showed satisfactory reliability. Effects of driving behaviour and personality factors on number of accidents were analysed by using regression analysis with a Poisson or negative binomial distribution.

The main results were rather consistent in all three data sets. Ordinary non-aggressive driving violations (e.g. speeding, dangerous overtaking, tailgating) predicted accidents in all three data sets. Effects of personality factors on accident involvement were mostly mediated by ordinary (studies 1,2 and 3) or aggressive (study 3) driving violations, although some personality factors also had a direct effect on the number of accidents. Driver errors did not predict the number of accidents. These results support the earlier findings that driving style rather than driving skills determine how safe a driver is. Hence, personality characteristics and attitudes influence driving style which, in turn, is related to accident liability.

In April 2000, the 1^st Annual Scholarship has been made to:

on the subject of:

200 Romanian, Hungarian, German, and Gypsy, preponderantly middle-income, parents of 9-11 year old children, all from Timisoara (Romania), were interviewed regarding goals, practices and difficulties of child-rearing. Their beliefs were compared with the American parents' beliefs. Parents' education level was considered.

Results: The most important goal for all parents involved in the study was "providing consistency and a loving, secure environment" for their children. A cross-cultural difference in American-European parental child-rearing views were the importance of maintaining their family values as against the building of their child's individuality. European parents appreciated the goal "instilling family values" as more important than American parents who, by contrast, thought that a more important goal in a child-rearing goals' hierarchy was "building the child's self-esteem". That difference can be a reflection of the general collectivist (Romanian) or individualistic (American) societal orientation.

A cross-cultural difference between Romanian, Hungarian, German, and Gypsy parents' views was that Romanian and Gypsy child-rearing beliefs corresponded mostly to the collectivist society type, but Hungarian and German child-rearing beliefs were appropriate for the individualistic society type.

16 types of child-rearing practices (strategies) were identified. The most used strategies were those based on verbal methods: explanation, instruction and demonstration. There was also some coherence between child-rearing goals and child-rearing practices. A cross-cultural analysis revealed that there were more similarities than differences between Romanian, Hungarian, and German parents' reported practices. Less educated parents more often preferred using the "control" strategy, but better educated parents appreciated the strategies that supported the child's independence.

6 types of child-rearing difficulties were identified. Most often mentioned categories of difficulties were "child's personality features", inappropriate "social ambience", and "lack of time". Cross-cultural analysis of the reported difficulties revealed only few differences. For example, one of the differences was that Hungarian and German parents stressed more often than Romanian parents an inappropriate social ambience as a child-rearing difficulty. Less educated parents more often associated their problems of child-rearing with financial problems, and better educated parents more often perceived societal ambience as the initial source of the child-rearing problems.

Parents and teachers were asked to evaluate their children's school performance and level of social adaptation. Results revealed that children's school performance was better if they were monolingual (Romanian); completed a kindergarten' s education; their parents were Romanians, Orthodox, with college/professional educational degree and were employed at skilled/professional level. Children' s school adaptation was better if they were the only children in their families; monolingual (Romanian), and completed a kindergarten's education; their parents were Romanians, Orthodox, with at least high/technical school degree, worked (were not unemployed), and had only 1-2 children.