The development of antiretroviral therapy (ART) has dramatically improved survival for those living with human immunodeficiency virus (HIV), but whether ART improves cognitive functioning remains unclear. The aim of the present review was to examine systematically the extent to which ART improves cognition among individuals with HIV using meta-analytic methods. Twenty-three studies were included in the quantitative review. ART was associated with modest improvements in attention (mean d = .17; p < .001; 95% confidence interval [CI], .09/.25), executive function (mean d = .18; p < .001; 95% CI, .10/.26), and motor function (mean d = .24; p < .001; 95% CI, .16/.32). ART did not improve language, verbal memory, visual memory or visuospatial function. The extent to which cognition improved was correlated with the change in CD4 cell count following ART, suggesting a link between cognitive outcome and immune system integrity. Together, the present findings indicate that the neuropsychological test performance of most HIV patients taking ART is comparable to those not taking ART. Development of pharmaceutical treatments and rehabilitation strategies that target the cognitive effects of HIV infection is needed. (JINS, 2011, 17, 956–969)

A growing number of schools have increasingly de-emphasized the importance of providing physical activity opportunities during the school day, despite emerging research that illustrates the deleterious relationship between low levels of aerobic fitness and neurocognition in children. Accordingly, a brief review of studies that link fitness-related differences in brain structure and brain function to cognitive abilities is provided herein. Overall, the extant literature suggests that childhood aerobic fitness is associated with higher levels of cognition and differences in regional brain structure and function. Indeed, it has recently been found that aerobic fitness level even predicts cognition over time. Given the paucity of work in this area, several avenues for future investigations are also highlighted. (JINS, 2011, 17, 975–985)

Multiple sclerosis (MS) is an inflammatory disease of the central nervous system, resulting in physical, cognitive and affective disturbances, with notable declines in the ability to learn and retain new information. In this study, we examined if higher levels of physical activity in MS individuals were associated with an increased resting-state connectivity of the hippocampus and cortex, resulting in better performance on a task of episodic memory. Forty-five individuals with a clinically definite diagnosis of MS were recruited for the study. Consistent with previous reports, hippocampus was functionally connected to the posteromedial cortex, parahippocampal gyrus, superior frontal gyrus, and the medial frontal cortex. Higher levels of physical activity in MS patients were associated with an increased coherence between the hippocampus and the posteromedial cortex (PMC). The increased connectivity between these two regions, in turn, was predictive of better relational memory, such that MS patients who showed an increased coherence between the left (not right) hippocampus and the PMC also showed better relational memory. Results of the study are interpreted in light of the challenge of disentangling effects of physical activity from effects of disease severity and its neuropathological correlates. (JINS, 2011, 17, 986–997)

We examined the association of social activity with cognitive decline in 1138 persons without dementia at baseline with a mean age of 79.6 (SD = 7.5) who were followed for up to 12 years (mean = 5.2; SD = 2.8). Using mixed models adjusted for age, sex, education, race, social network size, depression, chronic conditions, disability, neuroticism, extraversion, cognitive activity, and physical activity, more social activity was associated with less cognitive decline during average follow-up of 5.2 years (SD = 2.7). A one point increase in social activity score (range = 1–4.2; mean = 2.6; SD = 0.6) was associated with a 47% decrease in the rate of decline in global cognitive function (p < .001). The rate of global cognitive decline was reduced by an average of 70% in persons who were frequently socially active (score = 3.33, 90th percentile) compared to persons who were infrequently socially active (score = 1.83, 10th percentile). This association was similar across five domains of cognitive function. Sensitivity analyses revealed that individuals with the lowest levels of cognition or with mild cognitive impairment at baseline did not drive this relationship. These results confirm that more socially active older adults experience less cognitive decline in old age. (JINS, 2011, 17, 998–1005)

Participation in cognitively stimulating leisure activities such as crossword puzzles may delay onset of the memory decline in the preclinical stages of dementia, possibly via its effect on improving cognitive reserve. We followed 488 initially cognitively intact community residing individuals with clinical and cognitive assessments every 12–18 months in the Bronx Aging Study. We assessed the influence of crossword puzzle participation on the onset of accelerated memory decline as measured by the Buschke Selective Reminding Test in 101 individuals who developed incident dementia using a change point model. Crossword puzzle participation at baseline delayed onset of accelerated memory decline by 2.54 years. Inclusion of education or participation in other cognitively stimulating activities did not significantly add to the fit of the model beyond the effect of puzzles. Our findings show that late life crossword puzzle participation, independent of education, was associated with delayed onset of memory decline in persons who developed dementia. Given the wide availability and accessibility of crossword puzzles, their role in preventing cognitive decline should be validated in future clinical trials. (JINS, 2011, 17, 1006–1013)

Monoamines, such as serotonin, dopamine, and norepinephrine, play a crucial role in the regulation of emotion processing and mood. In this study, we investigated how polymorphisms of the serotonin transporter (5-HTT) and catechol-O-methyltransferase (COMT) influence emotion recognition abilities. We recruited 88 female undergraduate students and assessed 5-HTT genotype and the COMT Val158Met polymorphism. The subjects completed two computerized tasks: The Penn Emotion Recognition Test (ER40) and the Penn Emotion Acuity Test (PEAT). For the ER40, we found that s-allele carriers performed significantly worse in the recognition of happy faces, but did better in the recognition of fearful faces, compared with homozygous l-carriers of the 5-HTT gene. Neither 5-HTT nor COMT genotypes influenced the ability to discriminate between different intensities of sadness or happiness on the PEAT. Moreover, there was no significant interaction between the two polymorphisms in their effect on performance on the ER40 or the PEAT. (JINS, 2011, 17, 1014–1020)

Prior studies suggest that autism spectrum disorders (ASD) are associated with a domain-specific memory impairment for faces. The underlying cause of this problem and its relation to impaired visual scanning of faces—particularly of the eyes—remains to be determined. We recorded eye movements while 22 high-functioning ASD and 21 typically developing (TD) adolescents encoded and later recognized faces and objects from a single, nonsocial object category (electric fans). Relative to TD subjects, ASD individuals had poorer memory for faces, but not fans. Correlational analyses showed significant relationships between recognition memory and fixations. Eye tracking during encoding revealed that TD subjects made more fixations to faces than fans, whereas ASD individuals did not differ in number of fixations made to each stimulus type. Moreover, although both the TD and ASD groups showed a strong preference for fixating the eyes more than the mouth, the ASD subjects were less likely than TD subjects to scan regions of the face outside of the primary facial features (i.e., eyes, nose, and mouth). We concluded that ASD individuals have a domain-specific memory impairment for faces relative to mechanical objects and that this impairment may be related to abnormal scanning during encoding. (JINS, 2011, 17, 1021–1029)

The effect of age at injury on outcome after acquired brain injury (ABI) has been the subject of much debate. Many argue that young brains are relatively tolerant of injury. A contrasting viewpoint due to Hebb argues that greater system integrity may be required for the initial establishment of a function than for preservation of an already-established function. A neuro-computational model of cortical map formation was adapted to examine effects of focal and distributed injury at various stages of development. This neural network model requires a period of training during which it self-organizes to establish cortical maps. Injuries were simulated by lesioning the model at various stages of this process and network function was monitored as “development” progressed to completion. Lesion effects are greater for larger, earlier, and distributed (multifocal) lesions. The mature system is relatively robust, particularly to focal injury. Activities in recovering systems injured at an early stage show changes that emerge after an asymptomatic interval. Early injuries cause qualitative changes in system behavior that emerge after a delay during which the effects of the injury are latent. Functions that are incompletely established at the time of injury may be vulnerable particularly to multifocal injury. (JINS, 2011, 17, 1030–1038)

Although the relationship between education and cognitive status is well-known, evidence regarding whether education moderates the trajectory of cognitive change in late life is conflicting. Early studies suggested that higher levels of education attenuate cognitive decline. More recent studies using improved longitudinal methods have not found that education moderates decline. Fewer studies have explored whether education exerts different effects on longitudinal changes within different cognitive domains. In the present study, we analyzed data from 1014 participants in the Victoria Longitudinal Study to examine the effects of education on composite scores reflecting verbal processing speed, working memory, verbal fluency, and verbal episodic memory. Using linear growth models adjusted for age at enrollment (range, 54–95 years) and gender, we found that years of education (range, 6–20 years) was strongly related to cognitive level in all domains, particularly verbal fluency. However, education was not related to rates of change over time for any cognitive domain. Results were similar in individuals older or younger than 70 at baseline, and when education was dichotomized to reflect high or low attainment. In this large longitudinal cohort, education was related to cognitive performance but unrelated to cognitive decline, supporting the hypothesis of passive cognitive reserve with aging. (JINS, 2011, 17, 1039–1046)

The current study examined regional frontal lobe volumes based on functionally relevant subdivisions in contemporaneously recruited samples of boys and girls with and without attention-deficit/hyperactivity disorder (ADHD). Forty-four boys (21 ADHD, 23 control) and 42 girls (21 ADHD, 21 control), ages 8–13 years, participated. Sulcal–gyral landmarks were used to manually delimit functionally relevant regions within the frontal lobe: primary motor cortex, anterior cingulate, deep white matter, premotor regions [supplementary motor complex (SMC), frontal eye field, lateral premotor cortex (LPM)], and prefrontal cortex (PFC) regions [medial PFC, dorsolateral PFC (DLPFC), inferior PFC, lateral orbitofrontal cortex (OFC), and medial OFC]. Compared to sex-matched controls, boys and girls with ADHD showed reduced volumes (gray and white matter) in the left SMC. Conversely, girls (but not boys) with ADHD showed reduced gray matter volume in left LPM; while boys (but not girls) with ADHD showed reduced white matter volume in left medial PFC. Reduced left SMC gray matter volumes predicted increased go/no–go commission rate in children with ADHD. Reduced left LPM gray matter volumes predicted increased go/no–go variability, but only among girls with ADHD. Results highlight different patterns of anomalous frontal lobe development among boys and girls with ADHD beyond that detected by measuring whole lobar volumes. (JINS, 2011, 17, 1047–1057)

The objective of this study was to test the hypothesis that apathy and depression are dissociable in Parkinson disease (PD) by conducting a confirmatory factor analysis (CFA) of items from two commonly used mood scales. A total of 161 non-demented PD patients (age = 64.1; ± 8.4 years) were administered the Apathy Scale and the Beck Depression Inventory-II. Items were hypothesized to load onto four factors: (1) an apathy factor representing loss of motivation, (2) dysphoric mood factor representing sadness and negativity, (3) loss of interest/pleasure factor representing the features common to both apathy and depression, and (4) a somatic factor representing bodily complaints. Results indicated a good fit for the overall CFA model, χ2 (128, N = 146) = 194.9; p<.01. RMSEA was .060 (p = .16). The four-factor model was significantly better than all alternative nested models at p < .001, including an overarching single factor model, representing “depression.” Results support the concept that apathy and depression are discrete constructs. We suggest a “factor based” scoring of the Apathy Scale and Beck Depression Inventory-II that disentangles symptoms related to apathy, depression, overlapping symptoms, and somatic complaints. Such scoring may be important in providing useful information regarding differential treatment options. (JINS, 2011, 17, 1058–1066)

Our objectives were to examine cognitive outcomes for extremely preterm/extremely low birth weight (EPT/ELBW, gestational age <28 weeks and/or birth weight <1000 g) children in kindergarten and the associations of these outcomes with neonatal factors, early childhood neurodevelopmental impairment, and socioeconomic status (SES). The sample comprised a hospital-based 2001–2003 birth cohort of 148 EPT/ELBW children (mean birth weight 818 g; mean gestational age 26 weeks) and a comparison group of 111 term-born normal birth weight (NBW) classmate controls. Controlling for background factors, the EPT/ELBW group had pervasive deficits relative to the NBW group on a comprehensive test battery, with rates of cognitive deficits that were 3 to 6 times higher in the EPT/ELBW group. Deficits on a measure of response inhibition were found in 48% versus 10%, odds ratio (95% confidence interval) = 7.32 (3.32, 16.16), p < .001. Deficits on measures of executive function and motor and perceptual-motor abilities were found even when controlling for acquired verbal knowledge. Neonatal risk factors, early neurodevelopmental impairment, and lower SES were associated with higher rates of deficits within the EPT/ELBW group. The findings document both global and selective cognitive deficits in EPT/ELBW children at school entry and justify efforts at early identification and intervention. (JINS, 2011, 17, 1067–1079)

Relationships between cortical brain structure and neurocognitive functioning have been reported in schizophrenia, but findings are inconclusive, and only a few studies in bipolar disorder have addressed this issue. This is the first study to directly compare relationships between cortical thickness and surface area with neurocognitive functioning in patients with schizophrenia (n = 117) and bipolar disorder (n = 121) and healthy controls (n = 192). MRI scans were obtained, and regional cortical thickness and surface area measurements were analyzed for relationships with test scores from 6 neurocognitive domains. In the combined sample, cortical thickness in the right rostral anterior cingulate was inversely related to working memory, and cortical surface area in four frontal and temporal regions were positively related to neurocognitive functioning. A positive relationship between left transverse temporal thickness and processing speed was specific to schizophrenia. A negative relationship between right temporal pole thickness and working memory was specific to bipolar disorder. In conclusion, significant cortical structure/function relationships were found in a large sample of healthy controls and patients with schizophrenia or bipolar disorder. The differences that were found between schizophrenia and bipolar may indicate differential relationship patterns in the two disorders, which may be of relevance for understanding the underlying pathophysiology. (JINS, 2011, 17, 1080–1093)

Recent advances in the treatment of children with severe congenital heart defects, such as hypoplastic left heart syndrome (HLHS) and other forms of univentricular heart (UVH), have significantly improved their survival rates. However, these children are at risk for various neurodevelopmental deficits. The aim of the present study was to assess cognitive development, expressive language, and behavior in 30-month-old children with univentricular heart. The participants were 22 children with HLHS, 14 with UVH, and 41 healthy control subjects. The Bayley Scales of Infant Development II, MacArthur Communicative Development Inventories, and Child Behavior Checklist were used for assessments. The results revealed that children with HLHS exhibited a significantly lower mean mental development index, more delays in expressive language functions, and more behavioral problems than did the control children. Two children with HLHS (9%) had mental development indexes below 50, indicating significantly delayed performance. The children with UVH differed from the control children with respect to their lower mean mental development index. These findings suggest that at the age of 30 months, neurodevelopmental deficits are especially prevalent in children with HLHS. Thus, early developmental screening, intervention, and neuropsychological follow-up until school age is recommended particularly for the children with HLHS. (JINS, 2011, 17, 1094–1103)

White matter alterations, leukoaraiosis (LA) on structural MRI, are associated with cognitive deficits and increased risk of dementia. LA may also impact on subjective memory complaints in otherwise healthy older adults. Little is known about the interplay between LA memory complaints and cognition. We investigated cognitive phenotypes associated with LA in 42 non-demented older adults categorized as having subjective cognitive complaints with no objective cognitive impairment—the subjective cognitive impairment group (SCI; n = 12), amnesic mild cognitive impairment (aMCI; n = 20), or healthy controls (HC; n = 11). We measured LA severity on MRI with a 40-point visual rating scale. Controlling for age and Mini-Mental State Examination (MMSE) score, analyses revealed multiple between-group differences. Follow-up linear regression models investigating the underlying contributors to each clinic group's cognitive profile indicated that LA contributed to learning slope variance (after accounting for age and MMSE) but only for the SCI group. Although the SCI group showed a significantly steeper learning slope when compared to HC and aMCI, increasing LA severity negatively impacted this group's rate of learning. This, in conjunction with the significant contribution of age on SCI learning slope performance variance suggests that greater LA burden at a younger age may contribute to subtle changes in learning for individuals with subjective cognitive complaints. (JINS, 2011, 17, 1104–1112)

Information processing speed is frequently cited as the primary cognitive domain impacted by multiple sclerosis (MS) and is usually evaluated with reaction time (RT) or rapid serial processing (RSP) measures. The present study compared the efficacy of RT and RSP measures to distinguish between patients with MS (N = 42) and healthy controls (N = 40). The RT measure was patterned after the Computerized Tests of Information Processing and included measures of simple, choice, and semantic RT. The RSP measures consisted of the Symbol Digit Modalities Test (SDMT) and the Stroop Test. Substantial differences in information processing speed between patients and controls were found on all tests, with slightly larger effect sizes for RSP measures than RT measures and for the SDMT than the Stroop Test. Binary logistic regression analyses showed RSP measures performed better than RT measures at distinguishing patients from controls, and likewise, the SDMT score performed better than the scores derived from the Stroop Test. Results are discussed in the context of three effects associated with common measures of processing speed: complexity, compounding, and augmentation. (JINS, 2011, 17, 1113–1121)

Theory of Mind (ToM) is the ability to attribute independent mental states to self and others to explain and predict behavior. Impairment of ToM is well established in developmental pathologies. In neurological populations, investigation of ToM is still rare but data suggest that ToM impairment could contribute to behavioral and social disturbances. In addition to neurological signs, multiple sclerosis (MS) presents with disorders of cognition and behavior directly related to brain damage. The aim of this study was to assess ToM abilities and recognition of facial emotional expression in adults with MS. We compared 64 patients with relapsing MS and 30 matched healthy controls on three levels of ToM tasks, a facial emotion recognition task, and a neuropsychological assessment. MS patients performed significantly worse than controls in emotion recognition and all ToM tasks (p < .02). These deficits were not correlated with demographic variables or neuropsychological test performance. These findings underscore the importance of assessing ToM and facial recognition in MS, as dysfunction in these areas may impact upon social interaction and, thus, impair quality of life for both patients with MS and their families. (JINS, 2011, 17, 1122–1131)

A growing body of literature has documented evidence for emotion labeling (EL) deficits after traumatic brain injury (TBI); however, long-term effects of TBI on EL abilities, particularly among young children, are unclear. We investigated EL abilities and socio-emotional outcomes in 32 children with moderate–severe TBI, 23 with complicated-mild TBI, and 82 children with orthopedic injuries (OI), shortly after injury and at 18 months post-injury. All children were between 3:0 and 6:11 years of age at the time of injury. Repeated measures analyses indicated that all groups showed improved EL performance between acute and 18-month assessments, but that the moderate–severe TBI group improved at a slower rate than the OI group, so that the two groups showed significantly different performance at 18 months. Emotion labeling ability did not significantly contribute to the prediction of socio-emotional outcomes after controlling for pre-injury functioning. These results provide preliminary evidence of emerging EL deficits after early childhood TBI that are related to injury severity but that do not predict social and behavioral outcomes. (JINS, 2011, 17, 1132–1142)