Development and Aging

Transcription

1 Scandinavian Journal of Psychology, 2009, 50, DOI: /j x Development and Aging Personality a resource or risk for successful development LEA PULKKINEN Department of Psychology, University of Jyväskylä, Finland Pulkkinen, L. (2009). Personality a resource or risk for successful development. Scandinavian Journal of Psychology, 50, Personality as a resource or risk for development was discussed in the light of the results of the ongoing Finnish Jyväskylä Longitudinal Study of Personality and Social Development (FJYLS) which the author has conducted since 1968 when the participants (N = 369, b. 1959) were 8 years of age. A general hypothesis presented within a two-dimensional framework of self-control and activity was that the child s high self-control of emotions and behavior would be associated with adaptive behavior in adulthood. The results have provided evidence in support for and limitations to the hypothesis. High self-control was a resource and low self-control was a risk for development, but there were gender differences that came out consistently in variable-oriented and person-oriented analyses. A resource factor for male adult social functioning was, compared to females, a broader construct of selfcontrol in childhood covering both more passive (compliant) and active (constructive) behavior; only active well-controlled (constructive) behavior was a resource for female functioning. A risk factor for female functioning was low self-control combined with passivity (internalizing behavior), whereas low self-control combined with activity (externalizing behavior) was a risk factor for male functioning. Childhood self-control was not directly associated with adult psychological functioning such as well-being but indirectly through social functioning such as career development. Low self-control in childhood was an antecedent of criminal behavior but only for those male offenders who committed offences in adulthood, not for those who had limited their offending to adolescence. Key words: Personality, self-control, longitudinal study, social functioning, gender differences. Lea Pulkkinen, Department of Psychology, PO Box 35, University of Jyväskylä, Finland. Fax: ; lea.pulkkinen@ psyka.jyu.fi Personality is a construct that is used to refer to an individual s distinctive and characteristic patterns of behavior, emotion and thought. Knowledge about individual differences in personality and about human development has increased remarkably since the rise of empirical research in personality psychology a half century ago and the increase in long-term longitudinal studies on human development. One of these is the ongoing Finnish Jyväskylä Longitudinal Study of Personality and Social Development (FJYLS) which I began 40 years ago. The present paper gives an overview of the FJYLS, and presents its key results concerning the role of personality characteristics in human development. Developmental processes in the present paper are viewed in two directions: by following children s development to adulthood, and by looking for the roots of adult behaviors in childhood. THE FINNISH JYVÄSKYLÄ LONGITUDINAL STUDY OF PERSONALITY AND SOCIAL DEVELOPMENT The point of departure My first research findings as a doctoral student of psychology in the 1960s showed that aggressive behavior was characteristic of only a small proportion of children. I became interested in how non-aggressive children behave in provocative situations. Nonaggression was generally referred to in psychological literature as inhibition of aggression which was negatively evaluated because, from the psychodynamic point of view, inhibition of aggression was expected to lead to personality pathology. On the other hand, animal studies had shown that rats engaged in fight or flight. I was critical of these conceptions of the alternatives to aggression, since I thought that human beings were more capable of choosing their behavior. I speculated that children learn to regulate their emotional behaviors cognitively, considering situational factors and behavioral alternatives with their anticipated consequences (Pitkänen, 1969). It was a novel idea. At that time, emotions were understood as states; the study of emotional regulation became a focus of study much later, in the 1980s and 1990s. A few analyses of nonaggressive behaviors under the conditions of frustration by the end of 1960s included the study by Block and Martin (1955) on constructive behavior and the study by N. Feshbach and S. Feshbach (1969) on empathy as an alternative to aggression. Psychological research was oriented to psychopathology, anxiety and aggression rather than to aspects of healthy functioning and prosocial behavior. Prosocial behavior gradually emerged as a focus of study in the 1970s and 1980s, and positive psychology has grown as a movement in psychology only since the start of this millennium (Seligman & Csikszentmihalyi, 2000). To understand non-aggressive behaviors, I introduced a twodimensional impulse control model; the dimensions were called number of overt responses and control of behavior (Pitkänen, 1969). The model involved the idea that both the expression and inhibition of impulses can be controlled cognitively. This idea helped me to define three types of non-aggressive behav- Published by Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. ISSN

2 Scand J Psychol 50 (2009) Personality a resource or risk for successful development 603 iors: constructive, compliant, and anxious behavior. The model was built within a learning theoretical framework with the idea that different behavioral types (response habits) were learnt in early social interaction, particularly, with the parents, but that they varied intra-individually depending on situational factors. I expected that due to life experiences, certain response habits become stronger than others in an individual s repertoire. Empirical data were collected with 8-year-old children to confirm the model as part of my doctoral dissertation (Pitkänen, 1969). This was the start of the Finnish Jyväskylä Longitudinal Study of Personality and Social Development. Table 1. Major data waves of the Finnish Jyväskylä Longitudinal Study of Personality and Social Development (FJYLS) Year Age Participants Methods boys 173 girls boys 167 girls 77 boys 77 girls men 67 women men 155 women men 152 women men 134 women men 124 women (in progress) Peer nomination Teacher rating Personality inventories Peer nomination Teacher rating Subject interview Parental interview Subject interview Self-control inventories Mailed questionnaire Interview + inventories Mailed questionnaire Interview + inventories Mailed questionnaire Interview + inventories Medical examination Mailed questionnaire Interview + inventories Medical examination Data waves of the FJYLS The cross-sectional study transitioned into a long-term longitudinal study. Table 1 presents the stages of the FJYLS. The participants (born in 1959, average age 8.5 years), 196 boys and 173 girls, were drawn from second-grade pupils as a random sample of 12 school classes. Data were collected using teacher ratings and peer nominations (Pitkänen, 1969). The same methods were used when the participants were followed up at age 14. In 6 years the participants were spread out from 12 school classes to 78. Due to limited financial resources of the young researcher, interviews had to be limited to 154 pupils at age 14 (in 1974) and either of the parents, and to 135 participants when they were 19 to 20 years of age (Pulkkinen, 1982). At age 27, all participants of the original sample were traced for a follow-up study in 1986, and again at ages 36 (in 1995), 42 (in 2001), and 50 (in 2009) with the support of funding from the Academy of Finland. Each time a mailed Life Situation Questionnaire (LSQ1-4) was completed by more than 80% of the available sample. The LSQs consisted of about 200 items concerning education, work, family, leisure, alcohol consumption, life satisfaction, values, personal control over development, and future orientation. Before the interview, the participants filled out the Eysenck Personality Questionnaire (at age 27), and at ages 36 to 50, the Karolinska Scales of Personality and the Big Five Personality Inventory. The interview included the Identity Status Interview (ages 27 to 50), the Life History Calendar (at ages 42 and 50), and discussion about important domains in people s lives, such as marital relationships and working conditions. Since age 36, the participants were presented with several selfadministered questionnaires during the interview session (Pulkkinen, 2006; Pulkkinen, Feldt & Kokko, 2005). 1 Medical examinations with laboratory tests were conducted at ages 42 and 50. Relevant register data were also collected. Aims of the present paper and theoretical background The general hypothesis of the FJYLS about individual resource factors for development has been that the child s high self-control of emotions and behavior would be associated with adaptive behavior in adulthood. Conversely, the child s low self-control was expected to be a risk factor for development and associated with maladaptive behavior in adulthood. In this article, evidence is presented in support for and limitations to these hypotheses considering gender differences and different adult outcomes. In the analysis of continuity in developmental paths from childhood to adulthood, more emphasis has been laid on heterotypic continuity referring to phenotypically different but theoretically connected behaviors (Kagan, 1971) than to homotypic continuity which refers to associations between phenotypically similar behaviors at different ages. Considering the long life span, from age 8 to 42 (analysis of age 50 data is still in progress), the assessment of phenotypically similar behaviors at different ages is rarely possible, although we have found homotypic continuity in aggression from childhood to middle age for both genders in Finland and the United States (Kokko, Pulkkinen, Huesmann, Dubow & Boxer, 2009). The framework of the interpretation of connections between behaviors stems from the impulse control model and personality traits. I included the construct of self-control into the impulse control model (Pulkkinen, 1982) and relabeled it as the model of emotional and behavioral regulation, along with the emerging study of emotion regulation in psychology in the 1990s (Pulkkinen, 1995). I use the concepts of self-control and self-regulation interchangeably, as it is commonly done (Vohs & Baumeister, 2004). The construct of self-control was introduced to the psychological literature earlier than self-regulation, and they have different histories as reviewed by Pulkkinen and Pitkänen (2009). Both constructs mean that a person exerts control over his or her responses (thoughts, emotions, impulses, performances) so as to pursue goals and live up to standards (ideals, targets, expectations of other people) (Peterson & Seligman, 2004). Self-control may result in emotion regulation that refers to the redirection, control, and modification of an emotional state, which enables an individual to

3 604 L. Pulkkinen Scand J Psychol 50 (2009) function adaptively (Eisenberg, Fabes, Guthrie & Reiser, 2002). Empirical research shows, for instance, that children s poor ability to regulate their attentiveness is associated with behavior problems through deficits in the regulation of negative emotions, whereas high effortful control is associated with positive emotional, social, and cognitive development (Eisenberg, Smith, Sadovsky & Spinrad, 2004). The model (Fig. 1) explicates that active but emotionally dysregulated behavior may appear as aggressiveness and other externalizing symptoms (Type I), whereas active but emotionally regulated behavior may appear as socially responsible, constructive behavior (Type II). Correspondingly, passive but emotionally regulated behavior may appear as prudent and compliant behavior (Type III), and passive but emotionally dysregulated behavior may appear as anxiety and other internalizing symptoms (Type IV). Externalizing and internalizing behaviors have low self-control in common. In terms of Eysenck s (1960) two temperamental factors of personality, the vertical axis of the model (activity) was expected (Pitkänen, 1969) to be associated with extraversion vs. introversion, and the horizontal axis (low self-control) to neuroticism vs. stability. H. J. Eysenck and S. B. G. Eysenck (1963) identified the dual nature of extraversion, and made a distinction between social and impulsive extraversion. Within the twodimensional model, impulsive extraversion is less controlled than social extraversion. The different facets of extraversion, warmth, gregariousness, assertiveness, activity, excitement seeking, and positive emotions are also recognized in the five factor description of personality (McCrae & Costa, 2003). Extraverted children may have strong impulses that they must learn to regulate, as Shiner (2006) notes, and risks for unsuccessful regulation may be higher in males. Our findings have shown that extraversion when it is expressed as well-controlled, responsible and constructive behavior (Type II) in childhood is a resource for positive partner relationships (Kinnunen & Pulkkinen, 2003) and other positive outcomes such as career success in females (Pulkkinen, Ohranen & Tolvanen, 1999). On the contrary, extraversion when it is less controlled (impulsive and aggressive, Type I) in childhood is a risk for conflicting human relationships and predicts adult crime in men (Pulkkinen, Lyyra & Kokko, 2009). Neuroticism or high negative emotionality may also predict different outcomes depending on the emotionality involved. It may lead to internalizing symptoms such as depression if loaded by anxiety (Type IV), and to externalizing symptoms such as aggression (Type I) if loaded by irritability. For partner relationships, our results showed that women and men who had divorced by age 36 had been rated by their teachers at age 8 as more aggressive than women and men living in a marriage (Kinnunen & Pulkkinen, 2003). We also found that both women and men living in a poor marriage had been rated by their teachers as being more anxious. Thus low self-control was a risk in marriage in two ways: more aggressive individuals divorced and more anxious individuals remained in a poor relationship. Nevertheless neuroticism, particularly higher fearfulness, may also predispose children to positive outcomes because of their guilt-proneness and social anxiety (Shiner, 2006). Our results have confirmed this. Social anxiety in early adolescence protected males from heavy drinking in late adolescence and females from frequent drinking in adulthood (Pitkänen, Kokko, Lyyra & Pulkkinen, 2008). A parallelism of the two basic dimensions of the model can also be seen with Rothbart s (Rothbart & Derryberry, 1981) conception of temperament. She saw temperament as constitutionally Passive Prudent Tense Compliant III IV Anxious HIGH SELF- CONTROL Stable Constructive II I Labile Aggressive LOW SELF- CONTROL Responsible Impulsive Resource factors Active Risk factors Males Females SOCIAL ACTIVITY Males Females Fig. 1. A model of emotional and behavioral regulation, and resource and risk factors for further development in men and women.

4 Scand J Psychol 50 (2009) Personality a resource or risk for successful development 605 based differences in reactivity and self-regulation. A baby s early behavior is reactive to external stimuli and internal changes, but later self-regulatory systems, including inhibitory aspects of fear and the attentional flexibility of effortful control, develop to modulate reactivity. Effortful control has been defined as the ability to inhibit a dominant response to perform a subdominant response (Rothbart & Bates, 1998, p. 137). In the analyses of children s behavior, three factors with oblique rotation have been extracted (consistent with broad dimensions in adult behavior): surgency/extraversion including, for instance, high activity level, impulsivity, and positive anticipation; negative affectivity/neuroticism characterized by, for instance, fear and anger/frustration; and effortful control defined by, for instance, inhibitory control and (negatively) anger/frustration (Rothbart & Putnam, 2002). Effortful control was suggested to be developmentally related to the broad dimension of conscientiousness. Children s conscientiousness (close to Type II responsible and constructive behavior in Fig. 1) is associated with greater task-oriented self-control that promotes many positive outcomes, as reviewed by Shiner (2006). In our study, Type II behavior in childhood was associated with positive psychological functioning (well-being, coherence, self-esteem) in early middle age via educational and occupational achievements (Pulkkinen, Feldt & Kokko, 2006). Individuals who are high in conscientiousness both in childhood and adulthood are most likely to have good health at a later age (Martin, Friedman & Schwartz, 2007). The effects of conscientiousness on health are accumulating through health-enhancing behavior (Hampson, 2008). Our study has shown that adult conscientiousness is associated with calm mood and emotion regulation (Kokkonen & Pulkkinen, 2001), and that emotion regulation is associated with good subjective health (Kinnunen, Kokkonen, Kaprio & Pulkkinen, 2005). Effortful control or self-control is also involved in agreeableness, but as oriented to human relationships rather than to task performance; agreeableness predicts good peer relationships and school adjustment (Shiner, 2006). In our study, adult agreeableness was preceded by high childhood compliance (Type III behavior, Fig. 1) and low aggressiveness, and it was associated with a stable career line and low alcoholism and depression (Laursen, Pulkkinen & Adams, 2002). Thus many differences in personality can be interpreted as being associated with the quality of self-control. GENDER DIFFERENCES IN SELF-CONTROL AS A RESOURCE AND RISK FACTOR FOR DEVELOPMENT Formation of developmental paths The results of the FJYLS have generally confirmed the hypothesis about the role of self-control in the development of social functioning which refers to external criteria (e.g., education, stability of career line, socialization, controlled drinking) for adaptive behavior. Significant and unexpected gender differences have, however, emerged. They can be summarized as follows (cf. Fig. 1): (1) A resource factor for male adult social functioning is a broad construct of self-control in childhood covering both compliant and constructive (Type III and II) behaviors. (2) A resource factor for female social functioning is an active component of self-control in childhood indexed by constructiveness (Type II) and activity. (3) A risk factor for male adult social functioning is an active component of low self-control in childhood indexed by aggressive (Type I) behavior and activity. (4) A risk factor for female adult social functioning is a broad construct of low selfcontrol covering mainly passivity and anxious (Type IV) behavior. Several findings proved these conclusions. Both Type III (compliant) and II (constructive) behaviors in boys, but Type II behavior and activity in girls correlated positively with the length of education and stability of working career by age 27 (Pulkkinen, 1998). A path analysis (Pulkkinen et al., 1999) confirmed the gender difference. In men, compliant and constructive behaviors at ages 8 and 14 were linked to agreeableness at age 27, which was further associated with a high career orientation at age 36. On the contrary, constructive behavior associated with social activity at ages 8 and 14 in women was linked to extraversion at age 27, which was further associated with high career orientation at age 36. Furthermore, women who were staying in a stable and high quality relationship compared to poor quality, had been more constructive and active in childhood, while corresponding men had been more passive (Kinnunen & Pulkkinen, 2003). Correspondingly, Type I (aggressive) behavior in boys but Type IV (anxious) behavior and passivity in girls, correlated negatively with the length of education and career stability (Pulkkinen, 1998). A path analysis confirmed that girls social passivity associated with anxiety across the ages 8 and 14 was linked to Neuroticism at age 27, which was further associated with low career orientation at age 36 (Pulkkinen et al., 1999). In males, low self-control associated with aggressiveness was further associated with low career orientation. Gender differences were also found in the age 8 antecedents of problem drinking at age 42; for males problem drinking was preceded by low compliance and high activity and aggression, but for females by low constructiveness and low activity (Pitkänen et al., 2008). Roots of personality styles Gender differences also appeared in a person-oriented analysis where a hierarchical clustering technique (WARD) was used for extracting personality styles from variables assessed at age 27 representing personality characteristics, life orientation, and social functioning (Pulkkinen, 1996; Pulkkinen et al., 2005). One of the three clusters emerged for maladjustment, and it was gender-specific: the cluster was called Undercontrolled for males, and Brittle for females. Both the Brittles and the Undercontrolled were more neurotic, less agreeable, and less integrated into society than their adjusted same-sex counterparts, but the Brittles were more highly introverted and anxious (Type IV, cf. Fig. 1), whereas the Undercontrolled were more highly aggressive (Type I) than their same-sex counterparts. Similar differences between these clusters emerged already at age 8.

5 606 L. Pulkkinen Scand J Psychol 50 (2009) Two other clusters described more adjusted behavior, but there were also gender differences. Characteristics that indicated emotional and behavioral regulation were more highly influential organizing factors beyond the formation of adaptive styles in men than in women. The male clusters were called Resilient and Introverted (or Overcontrolled). Resilient men were socially active and well controlled, whereas the Introverted men were characterized by both compliance and anxiety. The Introverted men and the Brittle women had some commonalities, but the significant difference was that while the Brittle women had more problems than other women in social functioning, such as lower career stability and long-term unemployment, higher depression, and lower sense of coherence, the Introverted men did not differ from the other men in these respects; problems in social functioning differentiated the Undercontrolled men from the other men. Differences in female adaptive behavior were more notably formed on the basis of academic success and social roles than emotional and behavioral regulation. Adaptive but nonconventional (Individuated) women were oriented to studies, career, and culture, whereas the Traditionals were more family oriented. The only personality characteristic in which these two clusters differed at age 8 or at ages 27 to 42 was higher agreeableness of the Traditionals in adulthood. Differences in paths to social and psychosocial functioning In our study of successful development (Pulkkinen, Nygren & Kokko, 2002), outcome measures included both social functioning and psychological functioning at age 36; the latter refers to internal criteria of adaptive behavior, such as psychological well-being, self-esteem, and life satisfaction. Our hypothesis was that self-control in childhood, cognitive-motivational orientations at school, and home environment would explain both social and psychological functioning in adulthood. In the structural equation modeling (SEM), 75% of the variation of social functioning in men, and 52% in women were explained. Both constructive and compliant behavior (Types II III) in men, but only constructive behavior (Type II) in women contributed to the explanation. In the SEM model, psychological functioning was not explained by childhood factors. There was, however, a correlation between social and psychological functioning which indicated that there was dependence between them; causality could not be attributed to this relationship. In the study of antecedents of psychological functioning at age 42, we did not find significant correlations between childhood characteristics and psychological functioning, but we found that career success correlated with several indicators of psychological functioning. We formed a latent variable for psychological functioning of middle-aged people indexed by psychological well-being, self-esteem, social coherence, and trust in life, and two latent factors for career success, one for the level of career, including the length of education and occupational status, and another for career stability (Pulkkinen et al., 2006). Results based on SEM showed for males that adaptive behavior at ages 8 and 14 covering both Type II and III behaviors explained both components of career success which, on the other hand, accounted for psychological functioning. For females, links were more specific. Only Type II behavior explained one component of career success, namely career level, which accounted for psychological functioning. As regards risk factors, we have found that childhood aggression may result in a cycle of maladaptation at age 14 indicated by poor school success, lack of interest in school work, punishments at school, and truancy which, on the other hand, is linked to long-term unemployment both directly and indirectly via lack of occupational alternatives or heavy drinking (Kokko & Pulkkinen, 2000). Long-term unemployment, on the other hand, increases the risk of psychological distress (Kokko, Pulkkinen & Puustinen, 2000). High aggression is a risk factor for development, but the risk is smaller if the child exhibits some prosocial (Type II) behavior (Kokko & Pulkkinen, 2000). We have also found that children who are low in Type II behavior are at risk for delinquent behavior even when they do not display externalizing problems (Hämäläinen & Pulkkinen, 1996). ROOTS OF ADULT BEHAVIOR Criminal behavior In addition to using structural equation modeling for analyzing developmental pathways, we have searched for roots of adult criminal and intemperate behaviors in childhood behaviors. There are different theories about crime, and most of them emphasize social factors such as poverty and low social control as sources of crime. An exception is the general theory of crime by Gottfredson and Hirschi (1990) which argues that individuals possessing high self-control would be substantially less likely at all periods of life to engage in criminal acts (p. 89). Individuals who lack self-control tend to be, for instance, impulsive, insensitive, risk-taking, and short-sighted, and they will tend therefore to engage in criminal and analogous acts (p. 90). They assumed that low self-control can be identified prior to the age of responsibility for crime, and it tends to persist through life. In our study, self-control was identified prior to the age of responsibility for crime which made it possible to test these assumptions. Four groups of male offenders were formed: adolescence-limited offenders who ended offending by age 21; persistent offenders who were offending in adolescence and adulthood; adult-onset offenders whose offending began after age 20; and non-offenders who had not reported any offences or had not been registered for any crime (Pulkkinen et al., 2009). Offender groups could not be formed for females due to their lower frequency of offending. Differences between male offender groups in self-control indicated by aggressiveness and compliance (Type I and III behaviors) emerged at age 8 and 14, but consistently only between the non-offenders and persistent offenders. The only characteristic which differentiated the non-offenders from all offender groups was higher social passivity at age 14 (not yet at

6 Scand J Psychol 50 (2009) Personality a resource or risk for successful development 607 age 8). In adulthood, however, the non-offenders had higher self-control compared to all offenders as indicated by lower aggressiveness and neuroticism. The non-offenders also had less psychosomatic symptoms than all offenders. The adolescence-limited offenders did not differ from the non-offenders in childhood self-control, but in adolescence they were more peer-oriented and used more alcohol. In early middle age, they were controlling their drinking, and they were well established in terms of work and marriage. On the contrary, the adult-onset offenders displayed problem behaviors at early school age, but not in adolescence, and in adulthood they scored higher in risk-taking and alcoholism than the non-offenders. The assumption presented by Gottfredson and Hirschi (1990) on low self-control prior to committing crime was confirmed for persistent offenders and partially for adult-onset offenders, but not for adolescence-limited offenders. Temperance Another grouping of adults was made on the basis of temperance (Pulkkinen & Pitkänen, 2009). Temperance is one of six universal virtues (Peterson & Seligman, 2004), and it refers to strengths that protect against excess. The word temperance has also been used to refer to sobriety, abstaining from excess, often from alcohol. Our hypothesis was that drinking, smoking, excessive weight, and norm breaking would be interrelated in a way that would describe interindividual differences in temperance, and that these differences would be associated with individuals personality, particularly, self-control. Three groups were formed, separately for each gender, using data collected in adulthood (from age 27 to age 42). In the first group, called Temperate, criteria were: intoxicated less often than monthly, no smoking, and no registered crime after age 20. For overweight, criteria were: body mass index, BMI > 25 for women and BMI > 30 for men. In the intermediate group, individuals were between the first and third group in drinking. In the third group, called Intemperate, men had to meet at least three criteria and women at least two criteria among the following: intoxicated at least once a week, regular smoking, registered for crime after age 20, and BMI > 30. Because women were, on average, more temperate than men in all criteria, cutoff points had to be compromised. The comparison of the groups revealed that the temperate men and women had been higher in teacher rated self-control indexed by constructive and compliant behavior (men already at age 8), and lower in aggression (men already at age 8). The temperates had had better school success indicated by great point average, and longer education. The intemperates had displayed more conduct problems at age 14 than the temperates, and they had started drinking at a younger age than the temperates. Consequently, they had more problems associated with drinking behavior in adulthood. Being overweight at age 14 in terms of BMI was more common among the intemperates compared to the temperates. The results concur to other findings with FJYLS data: first, if drinking was started at age 14 or younger, the risk for alcoholism was fourfold compared to the age of onset of drinking at age 18 or later (Pitkänen, Lyyra & Pulkkinen, 2005); and second, individuals who weighed more than the others in childhood were more likely to be obese than to have normal weight (Kinnunen, Pietiläinen & Rissanen, 2006). Reasons for continuity in behavior Reasons for continuity in self-control lay in individuals innate characteristics and significant life circumstances as shown by a Finnish twin study with 12-year-old twins: 45 56% of the variation of Type II and III behaviors (cf. Fig. 1) were explained by additive genetic factors (A), one-third by common environmental factors (C), and one-sixth by unique environmental factors (E) (Vierikko, Pulkkinen & Rose, 2006). The percentages were very similar for low self-control indicated by hyperactivityimpulsivity and inattention, but for social anxiety (Type IV), the explanation of genetic factors was 82% in males and 60% in females, while C was 0 and 24%, respectively. For aggressiveness (Type I), genetic effects were, however, lower in males (27%) than in females (54%), while common environmental effects were higher in males; 66 and 37%, respectively (Vierikko, Pulkkinen, Kaprio & Rose, 2004). Thus, genetic factors explained more variance in the expression of low self-control that was less gender-typical or risky for future development (anxiety in males, aggression in females), while common environmental factors contributed relatively more to risky behaviors (aggression in males, anxiety in females). Environmental factors include physical and social environment. Physical environment such as noise, toxins, and maternal smoking and drinking during pregnancy may have negative effects on children s cognitive functions and socioemotional behavior (Evans, 2006). More attention has, however, been paid to social environment, particularly socioeconomic status. Nevertheless, the socioeconomic status of the family of origin has played a negligible role in the formation of individuals life paths in our Finnish study; more differentiating factors have been found in human relationships at home when the child was in preadolescence. It is to be noted that many aspects of home environment are confounded by genetic factors. Successful development has been associated with (1) good relationship between the parents; (2) the preadolescent child s good relationship with the father; (3) the mother s parenting with support, consideration of the child s opinion, supervision, and guidance; and (4) lack of physical punishment (Kokko & Pulkkinen, 2000; Pitkänen et al., 2008; Pulkkinen et al., 2002). Furthermore, our results have shown that controlled parental drinking is associated with positive development in children. Parental drinking was linked, particularly, to sons problem drinking in middle age, while maternal smoking was linked to daughters problem drinking in middle age (Pitkänen et al., 2008). It was common to both genders that the parents of the intemperate individuals had smoked cigarettes and used alcohol more than the parents of the temperates (Pulkkinen & Pitkänen, 2009).

7 608 L. Pulkkinen Scand J Psychol 50 (2009) Social environment covers human relationships also in other settings such as at school. Our results have shown that school years shape the development of self-control, and it may have long-term consequences. In females particularly, low self-control at age 14 was more predictive of negative outcomes such as middle age problem drinking than low self-control at age 8 (Pitkänen et al., 2008). For positive outcomes it is notable that continuity in male prosocial (Type II) behaviors over school years was positively associated with educational status at age 27 that explained occupational status at age 42 (Dubow, Huesmann, Boxer, Pulkkinen & Kokko, 2006). Interventions at school would be important for the reduction of the escalation of low self-control and for promoting high self-control, as proposed by Pulkkinen (2004). CONCLUSIONS The results of the FJYLS have provided evidence in support for and limitations to the hypothesis that low self-control is a risk factor and high self-control is a resource for further development. Limitations concerned, first, gender differences that came out consistently in variable-centered and person-centered analyses. High self-control in childhood was generally a resource for and low self-control a risk for adult development, but activity associated with it involved more risks (externalizing symptoms) for males (Type I; cf. Fig. 1) and passivity (internalizing symptoms) for females (Type IV). In individuals, personality characteristics such as activity and self-control appear in various combinations or profiles. If interpreted in terms of personality traits the results suggest that in males and females different personality profiles containing certain facets of neuroticism and extraversion/introversion are associated with different and distinct developmental outcomes for each gender.. Active well-controlled behavior (Type II) in childhood was a resource factor for the future in females, whereas well-controlled behavior in general (Types II and III) was a resource factor in males. The Type II and III behaviors are possibly associated with the personality traits of conscientiousness and agreeableness, respectively. These traits were not, however, measured in childhood in the present study, because they were not yet well-established personality dimensions in the late 1960s. Therefore, new longitudinal studies are needed for the verification of the suggestions that childhood conscientiousness and agreeableness would be associated with future development in a gender-specific way. Other limitations concerned developmental outcomes. First, our results showed that high self-control in childhood and preadolescence was not directly linked to psychological functioning with respect to internal criteria for adaptive behavior, although it was linked to positive social functioning in adulthood with respect to external criteria for adaptive behavior. High self-control was an antecedent of career success which was associated with psychological functioning. High self-control in childhood was also associated with other outcomes for social functioning in adulthood, such as marital stability and lower criminal behavior, that were associated with psychological well-being. Second, low self-control in childhood was an antecedent of criminal behavior only for those male offenders who committed offences in adulthood; most likely in those who had committed offences both in adolescence and adulthood. Men, who had limited their offending to adolescence, did not differ in self-control from non-offenders in childhood. In middle age, they were wellestablished in society. Third, the most intemperate middle-aged individuals among men and women had lower self-control in childhood than the most temperate individuals independent of the absolute criteria. Women exhibited more temperate behavior than men. Consequently, personality characteristics may either be risks or resources for successful development as a cumulative process. More research would be needed to understand environmental factors and confounding genetic factors contributing to the formation of and continuity in self-control. In spite of significant contributing factors, the individual is ultimately the agent of his or her life through personal choices that affect the direction of development. This paper presents a long-term longitudinal project that has been funded by the Academy of Finland since 1986, most recently from grant The project was a part of the Finnish Centre of Excellence Programme Human Development and Its Risk Factors (44858) in 1997 to Numerous researchers and students have participated in this long-term longitudinal study. I am thankful to all of them, and particularly, the present Co-Principal Investigators Katja Kokko, PhD, Research Fellow at the Academy of Finland, and Marja-Liisa Kinnunen, MD, PhD for their collaboration. NOTE 1 More information about the methods used in the FJYLS can be found in REFERENCES Block, J. & Martin, B. C. (1955). Predicting the behavior of children under frustration. Journal of Abnormal and Social Psychology, 51, Dubow, E. F., Huesmann, L. R., Boxer, P., Pulkkinen, L. & Kokko, K. (2006). Middle childhood and adolescent contextual and personal predictors of adult educational and occupational outcomes: A mediational model in two countries. Developmental Psychology, 42, Eisenberg, N., Fabes, R. A., Guthrie, I. K. & Reiser, M. (2002). The role of emotionality and regulation in children s social competence and adjustment. In L. Pulkkinen & A. Caspi (Eds.), Paths to successful development: Personality in the life course. (pp ). Cambridge, UK: Cambridge University Press. Eisenberg, N., Smith, C. L., Sadovsky, A. & Spinrad, T. L. (2004). Effortful control: Relations with emotion regulation, adjustment, and socialization in childhood. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, and application (pp ). New York: Guilford Press.

8 Scand J Psychol 50 (2009) Personality a resource or risk for successful development 609 Evans, G. W. (2006). Child development and the physical environment. Annual Review of Psychology, 57, Eysenck, H. J. (1960). The structure of human personality. London: Methuen. Eysenck, H. J. & Eysenck, S. B. G. (1963). On the dual nature of extraversion. British Journal of Social and Clinical Psychology, 2, Feshbach, N. & Feshbach, S. (1969). The relationship between empathy and aggression in two age groups. Developmental Psychology, 1, Gottfredson, M. R. & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press. Hämäläinen, M. & Pulkkinen, L. (1996). Problem behavior as a precursor of male criminality. Development and Psychopathology, 8, Hampson, S. E. (2008). Mechanisms by which childhood personality traits influence adult wellbeing. Current Directions in Psychological Science, 17, Kagan, J. (1971). Change and continuity in infancy. New York: Wiley. Kinnunen, M.-L., Kokkonen, M., Kaprio, J. & Pulkkinen, L. (2005). The associations of emotion regulation and dysregulation with the metabolic syndrome factor. Journal of Psychosomatic Research, 58, Kinnunen, M.-L., Pietiläinen, K. & Rissanen, A. (2006). Body size and overweight from birth to adulthood. In L. Pulkkinen, J. Kaprio & R. J. Rose (Eds.), Socioemotional development and health from adolescence to adulthood (ss ). Cambridge, UK: Cambridge University Press. Kinnunen, U. & Pulkkinen, L. (2003). Childhood socioemotional characteristics as antecedents of marital stability and quality. European Psychologist, 8, Kokko, K. & Pulkkinen, L. (2000). Aggression in childhood and long-term unemployment in adulthood: A cycle of maladaptation and some protective factors. Developmental Psychology, 36, Kokko, K., Pulkkinen, L., Huesmann, L. R., Dubow, E. F. & Boxer, P. (2009). Intensity of aggression in childhood as a predictor of different forms of adult aggression: A two-country (Finland and United States) analysis. Journal of Research on Adolescence, 19, Kokko, K., Pulkkinen, L. & Puustinen, M. (2000). Selection into longterm unemployment and its psychological consequences. International Journal of Behavioral Development, 24, Kokkonen, M. & Pulkkinen, L. (2001). Examination of the paths between personality, current mood, its evaluation, and emotion regulation. European Journal of Personality, 15, Laursen, B., Pulkkinen, L. & Adams, R. (2002). The antecedents and correlates of agreeableness in adulthood. Developmental Psychology, 38, McCrae, R. R. & Costa, P. T. Jr. (2003). Personality in adulthood: A five-factor theory perspective (2nd edn). New York: Guilford Press. Martin, L.R., Friedman, H. S. & Schwartz, J. E. (2007). Personality and mortality across the life span: The importance of conscientiousness as a biopsychosocial attribute. Health Psychology, 26, Peterson, C & Seligman, M. E. P. (2004). Character strengths and virtues: A handbook and classification. New York: Oxford University Press. Pitkänen, L. (1969). A descriptive model of aggression and nonaggression with applications to children s behaviour. (Jyväskylä Studies in Education, Psychology and Social Research, No. 19). Jyväskylä, Finland: University of Jyväskylä. Pitkänen, T., Kokko, K., Lyyra, A.-L. & Pulkkinen, L. (2008). A developmental approach to alcohol drinking behavior in adulthood: a follow-up study from age 8 to age 42. Addiction, 103(Suppl. 1), Pitkänen, T., Lyyra, A.-L. & Pulkkinen, L. (2005). Age of onset of drinking and the use of alcohol in adulthood: A follow-up study from age 8 42 for females and males. Addiction, 100, Pulkkinen, L. (1982). Self-control and continuity from childhood to late adolescence. In P. B. Baltes, O. G. Brim Jr (Eds.), Life-span development and behavior (Vol. 4, pp ). Orlando, FL: Academic Press. Pulkkinen, L. (1995). Behavioral precursors to accidents and resulting physical impairment. Child Development, 66, Pulkkinen, L. (1996). Female and male personality styles: A typological and developmental analysis. Journal of Personality and Social Psychology, 70, Pulkkinen, L. (1998). Levels of longitudinal data differing in complexity and the study of continuity in personality characteristics. In R. B. Cairns, L. R. Bergman & J. Kagan (Eds.), Methods and models for studying the individual (pp ). Beverly Hills, CA: Sage. Pulkkinen, L. (2004). A longitudinal study on social development as an impetus for school reform towards an integrated school day. European Psychologist, 9, Pulkkinen, L. (2006). The Jyväskylä Longitudinal Study of Personality and Social Development (JYLS). In L. Pulkkinen, J. Kaprio & R. J. Rose (Eds.), Socioemotional development and health from adolescence to adulthood (pp ). New York: Cambridge University Press. Pulkkinen, L., Feldt, T. & Kokko, K. (2005). Personality in young adulthood and functioning in middle age. In S. Willis & M. Martin (Eds.), Middle adulthood: A lifespan perspective (pp ). Thousand Oaks, Ca: Sage. Pulkkinen, L., Feldt, T. & Kokko, K. (2006). Adaptive behavior in childhood as an antecedent of psychological functioning in early middle age: Linkage via career orientation. Social Indicators Research, 77, Pulkkinen, L., Lyyra, A.-L. & Kokko, K. (2009). Life success of males on non-offender, adolescence-limited, persistent and adult-onset antisocial pathways: Follow-up from age 8 to 42. Aggressive Behavior, 35, Pulkkinen, L., Nygren, H. & Kokko, K. (2002). Successful development: Childhood antecedents of adaptive psychosocial functioning in adulthood. Journal of Adult Development, 9, Pulkkinen, L., Ohranen, M. & Tolvanen, A. (1999). Personality antecedents of career orientation and stability among women compared to men. Journal of Vocational Behavior, 54, Pulkkinen, L. & Pitkänen, T. (2009).Temperance and the strengths of personality: Evidence from a 35-year longitudinal study (Submitted for publication). Rothbart, M. & Bates, J. E. (1998). Temperament. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (pp ). New York: Wiley. Rothbart, M. & Derryberry, D. (1981). Development of individual differences in temperament. In M. E. Lamb & A. L. Brown (Eds.), Advances in developmental psychology (Vol. 1, pp ). Hillsdale, NJ: Erlbaum. Rothbart, M. & Putnam, S. P. (2002). Temperament and socialization. In L. Pulkkinen & A. Caspi (Eds.), Paths to successful development: Personality in the life course (pp ). Cambridge, UK: Cambridge University Press. Seligman, M. E. P. & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American Psychologist, 55, Shiner, R. L. (2006). Temperament and personality in childhood. In D. K. Mroczek & T. D. Little (Eds.), Handbook of personality development (ss ). Mahwah, NJ: Lawrence Erlbaum. Vierikko, E., Pulkkinen, L., Kaprio, J. & Rose, R. J. (2004). Genetic and environmental influences on the covariance between aggression

9 610 L. Pulkkinen Scand J Psychol 50 (2009) and hyperactivity-impulsivity as rated by teachers and parents. Twin Research, 7, Vierikko, E., Pulkkinen, L. & Rose, R. J. (2006). Genetic and environmental factors in girls and boys socioemotional behavior. In L. Pulkkinen, J. Kaprio & R. J. Rose (Eds.), Socioemotional development and health from adolescence to adulthood (ss ). Cambridge, UK: Cambridge University Press. Vohs, K. D. & Baumeister, R. F. (2004). Understanding self-regulation: An introduction. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, and application (pp. 1 9). New York: Guilford Press. Received 24 July 2009, accepted 29 July 2009