Neuropsychological deficits are considered endophenotypes for schizophrenia, because they are not only found in patients but also in many of their unaffected relatives, albeit in attenuated form. It is not yet clear which of these deficits in relatives are related to genetic or to environmental causes. We tested effects of inferred genetic liability for schizophrenia on neurocognitive variables to address this problem.

Twenty-eight patients with schizophrenia, 129 non-affected biological parents and 143 matched controls were assessed with an extensive neuropsychological test battery including tests of attention, memory, executive functioning and motor soft signs. Twenty-two parents had an ancestral history of schizophrenia and therefore were hypothesized to be more likely than their spouses without such a history (n = 17) to carry a genetic risk for schizophrenia.

Unaffected parents of schizophrenic patients showed significant deficits in a wide array of neuropsychological tasks and task domains. However, comparison of more likely and less likely carriers of illness-related genes showed specifically attentional and executive functioning, but not memory, to vary with degree of inferred genetic loading.

Attentional and executive (frontal) impairments vary with genetic loading for schizophrenia and can be considered true endophenotypes for this disorder. Consequently, these functions are particularly suited to evaluate the functional impact of candidate genes for schizophrenia in future studies.

Disturbances of the oculomotor system are promising endophenotypes for schizophrenia. In two separate studies, we examined antisaccade task performance, a measure of inhibitory control, in first degree relatives of schizophrenic patients (genetic risk without manifest disorder) and in clinical high risk subjects with symptoms suggestive of a prodromal phase of schizophrenia.

In the first study, 41 parents of schizophrenia patients and 22 controls were tested with with a prosaccade task and an antisaccade task. Parents were grouped into more likely, less likely, and indeterminate risk carriers. The second study involved 160 subjects clinically at risk for schizophrenia, 32 first episode schizophrenic patients, and 76 healthy controls.

In study 1 we found an increase of antisaccade latencies and error rates in parents of schizophrenics which varied with inferred genetic load, more likely gene carriers performing worst. In study 2, antisaccade performance varied with symptom load: subjects at risk with basic symptoms only were unimpaired, while at-risk subjects who had experienced brief psychotic episodes (BLIPS) showed deficits similar to first episode patients.

Reduced inhibitory control of oculomotor performance is associated with genetic loading for schizophrenia, and also with symptoms placing subjects at imminent risk of psychosis.

Implicit memories like consumption habits and conditioned reactions to drug-related stimuli are operational in addiction and relapse. The affective startle paradigm is an attractive tool for the measurement of the incentive salience of drug-related cues. We tested whether the stronger appetitive valence of drug cues, shown in two recent startle studies in smokers, does persist after prolonged abstinence, and may thus contribute to relapse.

We examined the auditory startle reflex magnitude of mildly deprived (4-6 hours) heavy smokers (n = 24), former smokers (n = 16, mean abstinence interval 18 months), and non-smokers (n = 24) while they viewed smoking-related scenes or standardized unpleasant, neutral and pleasant control scenes from the International Affective Picture System.

As expected, non-smokers showed no appetitive reactions toward smoking-cues. In smokers, smoking-cues had both appetitive implicit (startle suppression) and explicit (ratings for valence and craving) motivational effects, resembling those of pleasant scenes and differing from neutral and unpleasant scenes. This effect was more pronounced in smokers who later relapsed after a smoking cessation program, and in smokers consuming less than 20 cigarettes per day. Former smokers, despite reporting no craving and negative reactions to smoking cues, still showed evidence of implicit appetitive valence of these cues.

Nicotine addiction results in automatic appetitive reactions to drug-cues, which does not vanish after prolonged abstinence and which may thus contribute to relapses. Heavy smoking may result in a progressive internalization of smoking habits and a decline in reactivity towards external smoking-associated cues.

Psychosis is preceded by cognitive and physiological alterations. This may be useful in the risk assessment in subjects with putatively prodromal symptoms, and could contribute to better understand the temporal unfolding of the disease.

The early recognition and intervention program of the German Research Network on schizophrenia defines early and late prodromal stages according to psychopathological criteria. For concurrent and prospective validation of these risk stages, subjects undergo neurocognitive, electrophysiological and oculomotor assessments of putative vulnerability markers. About 125 early prodromal subjects (defined by the presence of basic symptoms, Klosterkoetter et al. 2001), and 90 late prodromal subjects (defined by attenuated positive symptoms or by brief occurrences of psychotic symptoms) have been assessed at inclusion.

As compared to psychiatrically healthy matched controls, late prodromals have significantly inferior verbal memory, verbal fluency, visual motor skills, and working memory. Impairments are qualitatively similar, but less pronounced in subjects in an early prodromal stage, with deficits of immediate verbal memory, verbal fluency and visuomotor performance being significant. Both groups show reduced auditory startle prepulse inhibition. Impairments are not correlated with depression and general distress scores, and are also largely independent of prodromal and attenuated positive symptoms. In early prodromals, global cognitive performance is related to the occurrence of psychotic symptoms during follow-up. Auditory P 300 is reduced in both prodromal groups, and predicts transitions to psychosis.

Neurocognitive and neurophysiological assessments validate and improve psychopathological risk assessment, and allow to disentangle stable vulnerability markers from indicators of imminent risk.

Funded by the German Federal Ministry for Education and Research BMBF (grant 01 GI 9934).

Patients with psychosis display the so-called ‘Jumping to Conclusions’ bias (JTC) – a tendency for hasty decision-making in probabilistic reasoning tasks. So far, only a few studies have evaluated the JTC bias in ‘at-risk mental state’ (ARMS) patients, specifically in ARMS samples fulfilling ‘ultra-high risk’ (UHR) criteria, thus not allowing for comparisons between different ARMS subgroups.

In the framework of the PREVENT (secondary prevention of schizophrenia) study, a JTC task was applied to 188 patients either fulfilling UHR criteria or presenting with cognitive basic symptoms (BS). Similar data were available for 30 healthy control participants matched for age, gender, education and premorbid verbal intelligence. ARMS patients were identified by the Structured Interview for Prodromal Symptoms (SIPS) and the Schizophrenia Proneness Instrument – Adult Version (SPI-A).

The mean number of draws to decision (DTD) significantly differed between ARM -subgroups: UHR patients made significantly less draws to make a decision than ARMS patients with only cognitive BS. Furthermore, UHR patients tended to fulfil behavioural criteria for JTC more often than BS patients. In a secondary analysis, ARMS patients were much hastier in their decision-making than controls. In patients, DTD was moderately associated with positive and negative symptoms as well as disorganization and excitement.

Our data indicate an enhanced JTC bias in the UHR group compared to ARMS patients with only cognitive BS. This underscores the importance of reasoning deficits within cognitive theories of the developing psychosis. Interactions with the liability to psychotic transitions and therapeutic interventions should be unravelled in longitudinal studies.

As physical activity may modify the effect of the apolipoprotein E (APOE) ε4 allele on the risk of dementia and Alzheimer's disease (AD) dementia, we tested for such a gene–environment interaction in a sample of general practice patients aged ⩾75 years.

Data were derived from follow-up waves I–IV of the longitudinal German study on Ageing, Cognition and Dementia in Primary Care Patients (AgeCoDe). The Kaplan–Meier survival method was used to estimate dementia- and AD-free survival times. Multivariable Cox regression was used to assess individual associations of APOE ε4 and physical activity with risk for dementia and AD, controlling for covariates. We tested for gene–environment interaction by calculating three indices of additive interaction.

Among the randomly selected sample of 6619 patients, 3327 (50.3%) individuals participated in the study at baseline and 2810 (42.5%) at follow-up I. Of the 2492 patients without dementia included at follow-up I, 278 developed dementia (184 AD) over the subsequent follow-up interval of 4.5 years. The presence of the APOE ε4 allele significantly increased and higher physical activity significantly decreased risk for dementia and AD. The co-presence of APOE ε4 with low physical activity was associated with higher risk for dementia and AD and shorter dementia- and AD-free survival time than the presence of APOE ε4 or low physical activity alone. Indices of interaction indicated no significant interaction between low physical activity and the APOE ε4 allele for general dementia risk, but a possible additive interaction for AD risk.

Physical activity even in late life may be effective in reducing conversion to dementia and AD or in delaying the onset of clinical manifestations. APOE ε4 carriers may particularly benefit from increasing physical activity with regard to their risk for AD.

Recent studies have provided strong evidence that variation in the gene neurocan (NCAN, rs1064395) is a common risk factor for bipolar disorder (BD) and schizophrenia. However, the possible relevance of NCAN variation to disease mechanisms in the human brain has not yet been explored. Thus, to identify a putative pathomechanism, we tested whether the risk allele has an influence on cortical thickness and folding in a well-characterized sample of patients with schizophrenia and healthy controls.

Sixty-three patients and 65 controls underwent T1-weighted magnetic resonance imaging (MRI) and were genotyped for the single nucleotide polymorphism (SNP) rs1064395. Folding and thickness were analysed on a node-by-node basis using a surface-based approach (FreeSurfer).

In patients, NCAN risk status (defined by AA and AG carriers) was found to be associated with higher folding in the right lateral occipital region and at a trend level for the left dorsolateral prefrontal cortex. Controls did not show any association (p > 0.05). For cortical thickness, there was no significant effect in either patients or controls.

This study is the first to describe an effect of the NCAN risk variant on brain structure. Our data show that the NCAN risk allele influences cortical folding in the occipital and prefrontal cortex, which may establish disease susceptibility during neurodevelopment. The findings suggest that NCAN is involved in visual processing and top-down cognitive functioning. Both major cognitive processes are known to be disturbed in schizophrenia. Moreover, our study reveals new evidence for a specific genetic influence on local cortical folding in schizophrenia.

Whether late-onset depression is a risk factor for or a prodrome of dementia remains unclear. We investigated the impact of depressive symptoms and early- v. late-onset depression on subsequent dementia in a cohort of elderly general-practitioner patients (n = 2663, mean age = 81.2 years).

Risk for subsequent dementia was estimated over three follow-ups (each 18 months apart) depending on history of depression, particularly age of depression onset, and current depressive symptoms using proportional hazard models. We also examined the additive prediction of incident dementia by depression beyond cognitive impairment.

An increase of dementia risk for higher age cut-offs of late-onset depression was found. In analyses controlling for age, sex, education, and apolipoprotein E4 genotype, we found that very late-onset depression (aged ⩾70 years) and current depressive symptoms separately predicted all-cause dementia. Combined very late-onset depression with current depressive symptoms was specifically predictive for later Alzheimer's disease (AD; adjusted hazard ratio 5.48, 95% confidence interval 2.41–12.46, p < 0.001). This association was still significant after controlling for cognitive measures, but further analyses suggested that it was mediated by subjective memory impairment with worries.

Depression might be a prodrome of AD but not of dementia of other aetiology as very late-onset depression in combination with current depressive symptoms, possibly emerging as a consequence of subjectively perceived worrisome cognitive deterioration, was most predictive. As depression parameters and subjective memory impairment predicted AD independently of objective cognition, clinicians should take this into account.