This issue features papers using functional imaging in psychiatric disorders linked with neuropsychology to explore underlying brain systems, together with other neuropsychological papers, and a related Invited Review by Nelson et al. (pp. 163–174) on neural changes which may underlie the developmental social changes of adolescence. Three papers deal with psychiatric disorders in childhood and adolescence. Two further papers report studies of suicidal behaviours and of depression. An additional paper from Barrett et al. (pp. 281–293) describes the prevalence and features of schizophrenia in an unusual cultural population isolate, the Iban of Sarawak.

Background. Many changes in social behavior take place during adolescence. Sexuality and romantic interests emerge during this time, and adolescents spend more time with peers and less time with parents and family. While such changes in social behavior have been well documented in the literature, relatively few neurophysiological explanations for these behavioral changes have been presented.

Method. In this article we selectively review studies documenting (a) the neuronal circuits that are dedicated to the processing of social information; (b) the changes in social behavior that take place during adolescence; (c) developmental alterations in the adolescent brain; and (d) links between the emergence of mood and anxiety disorders in adolescence and changes in brain physiology occurring at that time.

Results. The convergence of evidence from this review indicates a relationship between development of brain physiology and developmental changes in social behavior. Specifically, the surge of gonadal steroids at puberty induces changes within the limbic system that alters the emotional attributions applied to social stimuli while the gradual maturation of the prefrontal cortex enables increasingly complex and controlled responses to social information.

Conclusions. Observed alterations in adolescent social behavior reflect developmental changes in the brain social information processing network. We further speculate that dysregulation of the social information processing network in this critical period may contribute to the onset of mood and anxiety disorders during adolescence.

Background. Dissociation, defined as a disruption in usually integrated mental functions, is found not only in DSM-IV dissociative disorders, but also in post-traumatic stress disorder and eating disorders. Dissociative phenomena are also common in the general population, and may reflect a constitutionally determined cognitive style rather than a pathological trait acquired through experiencing adverse life events. In pathological dissociation, evidence has been presented for episodic memory dysfunction. In contrast, in high-dissociative subjects increased performance has been found for episodic memory and dual task performance. These findings have been linked to changes in working memory capacity.

Method. In the present study, the authors sought to extend these findings by using functional magnetic resonance imaging during performance of two parametric working memory tasks. We tested 21 healthy low- and high-dissociative participants.

Results. High-dissociative participants performed slightly better during both tasks. Imaging data showed that both groups activated similar networks for both tasks, i.e. (bilateral) dorsolateral (DL) and ventrolateral prefrontal cortex (PFC), parietal cortex, and supplementary motor area. Group×task interactions were found in the high-dissociative group in L DLPFC and L parietal cortex; in the low-dissociative group in R fusiform gyrus. The differences in the high-dissociative group were independent from performance differences, implying that high-dissociative subjects generally recruit this network to a greater extent.

Conclusions. These results confirm earlier findings using a verbal WM task in high-dissociative participants, and are compatible with the conceptualization of non-pathological dissociation as an information-processing style, characterized by distinct attentional and mnemonic abilities.

Background. In a recent longitudinal study of first-episode schizophrenia patients, we found that while dysfunction of the right dorsolateral prefrontal cortex (DLPFC), right thalamus, left cerebellum and cingulate gyrus normalized with antipsychotic treatment and significant reduction in symptomatology, the left DLPFC, left thalamus, and right cerebellum remained disturbed. In the present study we investigated whether these abnormalities are also present in clinically stable, relatively well-functioning schizophrenia patients in comparison to control subjects during performance of the N-back working-memory task.

Method. Twelve schizophrenia and 12 control subjects completed the study. The functional images collected during scanning were analyzed using a random-effects model in a restricted set of six regions of interest (ROIs). In addition, the exploratory search in the entire brain volume was performed.

Results. The ROI analyses revealed relative underactivation in the region of the left DLPFC and the right cerebellum, as well as overactivation in the left cerebellum. The exploratory whole-brain search exposed additional overactivation in the medial frontal, anterior cingulate, and left parietal cortices.

Conclusions. The present study provides evidence of significant underactivations in stable schizophrenia patients in regions that we have previously observed to be dysfunctional in acutely psychotic and partially remitted patients, together with extensive overactivations in several regions that potentially reflect some compensatory mechanism or increased effort on the working-memory task.

Objective. Although studies have indicated that the human basal ganglia have a specific role in different memory systems, the functional significance of the striatal dopamine activities for the basal ganglia remains less clear. This study assessed the relationship between measures of striatal dopamine activities and indices of different memory systems in healthy individuals.

Method. Single photon emission computed tomography and [123I]IBZM (iodobenzamide) were used to assess the striatal dopamine D2/D3 receptor density in 62 healthy volunteers aged between 19 and 61 years. All subjects underwent a Wechsler Memory Scale – Revised test.

Results. Dopamine D2/D3 receptor densities in the striatum decline with age. Stepwise regression analysis showed that verbal delayed recall and working memory account for most of the variance in dopamine D2/D3 measurements. These relationships remain significantly after controlling for age effects.

Conclusions. Brain striatal dopamine activities are also significantly associated with various memory systems, in addition to motor functions. This may explain why patients with neuropsychiatric disorders may have both memory and motor impairments.

Background. Euthymic bipolar-I disorder (BP I) patients and their siblings have shown impairments in verbal learning and memory functions compared with controls, suggesting that these impairments may be genetic in origin. Reduced information-processing speed has been associated with impaired memory in the elderly, and recently in schizophrenia. The authors compared verbal learning and memory functioning in twins with BP I and co-twins to control twins, and examined whether the observed deficits are related to information-processing speed.

Method. Finnish Medical and Population Registers and Twin Cohorts were used to identify the BP I and control twins. Neuropsychological tests assessing verbal learning and memory, working memory, facial recognition, visual memory, and information-processing speed were administered to 26 BP I twins, 19 non-bipolar co-twins, and 114 controls. Group differences were analyzed by generalized estimation equation modeling.

Results. BP I patients, but not co-twins, showed impairments in all memory tests compared with controls. Female co-twins showed impairment in verbal learning and memory. Information-processing speed had a significant effect on encoding and learning efficiency.

Conclusions. This study showed for the first time that information-processing speed is related to memory functioning and verbal learning in BP I in a population-based, representative and euthymic sample. Furthermore, the data support the view that defects in verbal memory may be related to the genetic factors predisposing to BP I in females.

Background. The loss of a spouse has been found to have a negative effect on physical and mental health and leads to increased mortality. Whether conjugal bereavement also affects memory functioning has largely been unexamined. The present study investigates the effect of widowhood on memory functioning in older persons.

Method. The sample consisted of 474 married women and 690 married men aged 60–85 years in 1992, followed up in 1995 and 1998. During the study 135 (28%) of the women and 69 (10%) of the men lost their spouse. Linear regression analysis was used to examine whether widowed men and women differed from those who had not been widowed in rate of memory change over 6 years. Cross-domain latent-change models were subsequently used to evaluate the extent to which changes in memory are related to changes in other domains of functioning that may be affected by widowhood.

Results. Older adults who lost a spouse during follow-up showed a greater decline in memory over 6 years than those who remained married. A higher level of depressive symptoms at baseline was related to lower levels of memory functioning and a greater decline. Memory decline was unrelated to changes in depressive symptoms and physical health.

Conclusions. Loss of the spouse is related to a greater decline in memory in older adults. The absence of an association with physical functioning and the weak association with mental functioning suggest that losing a spouse has an independent effect on memory functioning.

Objective. The goal of this study is to determine the association between panic attacks in adolescence and the risk of personality disorders during young adulthood.

Method. Data were drawn from the Children in the Community Study, a longitudinal epidemiological study of psychopathology across the life-course in 717 individuals in the community. Multiple logistic regression analyses were used to determine the association between panic attacks during adolescence in 1983 and the risk of personality disorders during young adulthood in 1993, adjusting for differences in sociodemographic characteristics, adolescent personality disorders, and co-morbid depressive and substance use disorders.

Results. Panic attacks during adolescence (in 1983) were associated with an increased risk of any DSM-IV personality disorder (in 1993) during young adulthood, which persisted after adjusting for differences in sociodemographic characteristics, adolescent personality disorders, and co-morbid depressive and substance use disorders. Panic attacks were associated with a statistically significantly increased risk of Cluster A, B, and C personality disorders.

Conclusions. These data provide initial evidence that panic attacks early in life are a marker or risk factor for the development of personality disorders in young adulthood. Replication of these findings is needed, as is more in-depth investigation into the mechanism of this link. If replicated in future research, these results may reveal a novel potential pathway for identifying youth at high risk for personality disorders.

Background. Age at onset (AAO) has been useful to explore the clinical, neurobiological and genetic heterogeneity of obsessive–compulsive disorder (OCD). However, none of the various thresholds of AAO used in previous studies have been validated, and it remains an unproven notion that AAO is a marker for different subtypes of OCD. If AAO is a clinical indicator of different biological subtypes, then subgroups based on distinct AAOs should have separate normal distributions as well as different clinical characteristics.

Method. Admixture analysis was used to determine the best-fitting model for the observed AAO of 161 OCD patients.

Results. The observed distribution of AAO in OCD is a mixture of two Gaussian distributions with mean ages of 11·1±4·1 and 23·5±11·1 years. The first distribution, defined by early-onset OCD, had increased frequency of Tourette's syndrome and increased family history of OCD. The second distribution, defined by late-onset OCD, showed elevated prevalence of general anxiety disorder and major depressive disorder.

Conclusions. These results, based on a statistically validated AAO cut-off and those of previous studies on AAO in OCD, suggest that AAO is a crucial phenotypic characteristic in understanding the genetic basis of this disorder.

Background. A self-report screening scale of adult attention-deficit/hyperactivity disorder (ADHD), the World Health Organization (WHO) Adult ADHD Self-Report Scale (ASRS) was developed in conjunction with revision of the WHO Composite International Diagnostic Interview (CIDI). The current report presents data on concordance of the ASRS and of a short-form ASRS screener with blind clinical diagnoses in a community sample.

Method. The ASRS includes 18 questions about frequency of recent DSM-IV Criterion A symptoms of adult ADHD. The ASRS screener consists of six out of these 18 questions that were selected based on stepwise logistic regression to optimize concordance with the clinical classification. ASRS responses were compared to blind clinical ratings of DSM-IV adult ADHD in a sample of 154 respondents who previously participated in the US National Comorbidity Survey Replication (NCS-R), oversampling those who reported childhood ADHD and adult persistence.

Results. Each ASRS symptom measure was significantly related to the comparable clinical symptom rating, but varied substantially in concordance (Cohen's κ in the range 0·16–0·81). Optimal scoring to predict clinical syndrome classifications was to sum unweighted dichotomous responses across all 18 ASRS questions. However, because of the wide variation in symptom-level concordance, the unweighted six-question ASRS screener outperformed the unweighted 18-question ASRS in sensitivity (68·7% v. 56·3%), specificity (99·5% v. 98·3%), total classification accuracy (97·9% v. 96·2%), and κ (0·76 v. 0·58).

Conclusions. Clinical calibration in larger samples might show that a weighted version of the 18-question ASRS outperforms the six-question ASRS screener. Until that time, however, the unweighted screener should be preferred to the full ASRS, both in community surveys and in clinical outreach and case-finding initiatives.

Background. Suicidal behaviour and completed suicide are serious problems within British prisons, leading to significant morbidity and mortality, and are the focus of major efforts towards their prevention.

Aim. To explore the demographic, social and psychiatric correlates of suicidal behaviour in prisons in England and Wales and their relationship with health service use; and to develop a combined psychosocial model of risk.

Method. This report analyses the prevalence of suicidal ideation and suicide attempts in the ONS National Prison Survey, and their association with the presence of psychiatric disorders, personality disorder, substance abuse and social risk factors. These data were compared with data from the second national survey of psychiatric morbidity in adults living at home. In both surveys, a two-phase interviewing procedure was used, covering general health, health service use, assessment of psychiatric disorders, life events, social supports, suicidal behaviour, activities of daily living, sociodemographic data, substance abuse and intelligence.

Results. Suicidal thoughts and suicide attempts were commoner in prisons than in the general population and these were significantly associated with higher rates of psychosis, neurosis and personality disorder in prisons. In addition, demographic and factors such as being young, single, white, leaving school early and experiencing poor social support and significant social adversity were important risk factors for suicidal thoughts. Crucially, there was no separate category of people at suicidal risk who did not have psychiatric disorders.

Conclusions. The high rates of suicidal behaviour in prisons cannot be addressed without adequate attention to the high rates of psychiatric disorder and vulnerability factors in prisoners.

Background. In patients with chronic medical conditions, depression can be viewed as an expected reaction to illness or an independent condition requiring active treatment. We examine how clinical and functional outcomes of depression treatment compare in primary-care patients with and without chronic medical illness.

Method. Health plan administrative data were used to identify primary-care patients initiating antidepressant treatment, including a general sample (n=204) and cohorts with ischemic heart disease (n=68), diabetes (n=93), or obstructive lung disease (n=74). Telephone interviews at baseline, 2 months, and 6 months assessed depression (Structured Clinical Interview for DSM-IV, Hopkins Symptom Checklist depression scale), functional status (SF-36) and days of disability because of to illness.

Results. At baseline, depression severity in patients with diabetes and obstructive lung disease was similar to those without medical co-morbidity, but significantly lower in those with ischemic heart disease. Social and emotional functioning were similar across all groups, but those with chronic medical illness reported greater physical impairment. All groups showed significant improvement in depression over six months, but the rate of improvement was significantly slower in those with heart disease. All groups showed significant and similar improvement in social and emotional functioning, but physical functioning showed little change. Days of disability and restricted activity improved significantly in all groups. Improvement in depression during treatment was strongly related to change in disability.

Conclusions. Patients with chronic medical illness and co-morbid depression show significant improvements in mood, social and emotional functioning, and disability following initiation of depression treatment. Depression may be a stronger determinant of disability than is stable chronic medical illness.

Background. We present results of a study of treated rates of schizophrenia among the Iban of Sarawak, Malaysia. Most Iban live in longhouses, each comprising a kindred group of up to 300 individuals. Cultural practices such as minimal intermarriage with members of adjacent ethnic groups and in-depth genealogical knowledge make them a population suitable for genetic investigation. Iban culture is conducive to a focus on symptoms and illness, and to patterns of treatment-seeking behaviour that are enthusiastic and persistent.

Method. We identified all known cases of psychotic disorder within a defined catchment area based on an exhaustive survey of available medical records. From corresponding Malaysian census data (91056 persons), we report rates of treated schizophrenia in the Iban population, using three diagnostic systems, as well as the demographic and clinical characteristics of these individuals.

Results. The most frequent presenting complaints were insomnia and aggression. We found higher treated rates for narrowly defined schizophrenia among males, but no significant gender difference for age of onset. Estimates of treated rates to age 55 years (per 10000) for narrow schizophrenia were 41·9 (ICD-10), 56·5 (DSM-IV), and 83 (RDC), while the rates for broad schizophrenia were 105·5, 103·2, and 107·5 respectively.

Conclusions. Treated rates of schizophrenia were higher than the reported prevalence for many populations at risk, including many small-scale societies, although different methodological approaches may partly explain these findings. Given the cultural patterns of Iban treatment-seeking behaviour, treated rates of schizophrenia reported here may closely approximate the population prevalence of this disorder.

Brugha, T. S., Morgan, Z., Bebbington, P., Jenkins, R., Lewis, G., Farrell, M. & Meltzer, H. (2003). Social support networks and type of neurotic symptom among adults in British households. Psychological Medicine33, 307–318.

The authors have informed us that in this paper the 95% confidence interval for each odds ratio was miscalculated in analyses, the published estimates being wider than they should be. The terms Perceived Social Support and Primary Support Group 3 or Less, were reversed in Table 5 and in the Results section text. The main conclusion is substantially unaltered, which is that there is a higher risk of depression, depressive ideas and panic symptoms, than of any other neurotic symptom, in subjects with a Primary Support Group of three or less and in subjects with low perceived social support in all six logistic models. The revised Table 5 is published overleaf.