Major depressive disorder is often considered to be a homogenous disorder that changes in terms of severity; however, the presence of distinct subtypes and a variety of presenting symptoms suggests much heterogeneity. Aiming to better understand the relationship between heterogeneity and diagnosis we used an exploratory approach to identify subtypes of depression on the basis of clinical symptoms and neuropsychological performance. Cluster analysis identified two groups of patients distinguished by level of cognitive dysfunction with the more severe cluster being associated with melancholic depression. While the relationship between cluster and subtype was significant, only 58% of melancholic patients were assigned to cluster 1 (the more severe cluster) and 66% of non-melancholic patients assigned to cluster 2. Subtypes also displayed a distinctive profile of impairment such that melancholic patients (n = 65) displayed more variability in attention while non-melancholic patients (n = 59) displayed memory recall impairment. While melancholia and non-melancholia are associated with a more severe and less severe form of depression respectively, findings indicate that differences between melancholia and non-melancholia are more than simple variation on severity. In summary, findings provide support for the heterogeneity of depression. (JINS, 2012, 18, 361–369)

Mild traumatic brain injury (mTBI) affects a significant number of combat veterans returning from Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF). Although resolution of mTBI symptoms is expected over time, some individuals continue to report lingering cognitive difficulties. This study examined the contributions of self-reported mTBI injury characteristics (e.g., loss of consciousness, post-traumatic amnesia) and psychiatric symptoms to both subjective and objective cognitive functioning in a sample of 167 OEF/OIF veterans seen in a TBI clinic. Injury characteristics were not associated with performance on neuropsychological tests but were variably related to subjective ratings of cognitive functioning. Psychiatric symptoms were highly prevalent and fully mediated most of the relationships between injury characteristics and cognitive ratings. This indicates that mTBI characteristics such as longer time since injury and loss of consciousness or post-traumatic amnesia can lead to increased perceived cognitive deficits despite having no objective effects on cognitive performance. Psychiatric symptoms were associated with both cognitive ratings and neuropsychological performance, illustrating the important role that psychiatric treatment can potentially play in optimizing functioning. Finally, subjective cognitive ratings were not predictive of neuropsychological performance once psychiatric functioning was statistically controlled, suggesting that neuropsychological assessment provides valuable information that cannot be gleaned from self-report alone. (JINS, 2012, 18, 576–584)

Previous research has identified cognitive impairment in children with sickle cell anemia (SCA, Hemoglobin SS) compared with controls, partly accounted for by overt neuropathology after clinical stroke, “covert” (“silent”) infarction, and severity of anemia. However, cognitive deficits have also been identified in children with SCA with no history of stroke and a normal T2-weighted magnetic resonance imaging (MRI) scan. Our aim was to investigate whether nocturnal hemoglobin oxygen desaturation and sleep fragmentation could be associated with cognitive impairment in children with SCA. We assessed 10 children with SCA (9 with normal MRI) using neuropsychological measures of executive function. Cognitive assessment was immediately followed by overnight polysomnography to record nocturnal hemoglobin oxygen saturation and sleep arousals. Decreases in hemoglobin oxygen saturation and/or increased sleep arousals were associated with reduced performance on cognitive assessment. Nocturnal hemoglobin oxygen desaturation and sleep fragmentation may be a contributing factor to executive dysfunction in SCA. (JINS, 2012, 18, 168–173)

Cognitive changes are documented in bipolar disorder (BP). Cortical volume loss, especially in prefrontal regions, has also been reported, but associations between cognition and cortical abnormalities have not been fully documented. This study explores associations between cognitive performance and cortical parameters (area, thickness and volume) of the fronto-temporal cortex in 36 BP patients (25 BPI and 11 BPII). T1-weighted volumetric MRI images were obtained using a 1.5 Tesla scanner. Cortical parameters were measured using surface-based morphometry and their associations with estimated premorbid, current IQ, visual memory, and executive function explored. Premorbid IQ was associated with frontal cortical area and volume, but no such associations were present for current cognitive performance. Cortical parameters were not different in BPI and BPII patients, but the association between current IQ and temporal cortical area was stronger in BPII patients. The pattern of cortico-cognitive associations in BPI and BPII patients merits further consideration. (JINS, 2012, 18, 370–374)

The Iowa Gambling Task (IGT) is a measure of decision-making, in which alternative metrics have greater construct validity than conventional metrics. No large scale study has examined the neural correlates in healthy adults. We administered the IGT and structural MRI to 124 healthy participants. We analyzed the conventional IGT metric of advantageous minus disadvantageous choices (i.e., decks C + D minus decks A + B), and three alternative metrics based on choices from decks D and A alone, and all selections from each deck. Using regression and voxel-based morphometry, we examined regional gray matter volumes as predictors of IGT performance. No neural correlates of the conventional metric emerged, and the neural correlates of individual deck selections were disparate from one another. Alternative metrics showed expected neural correlates of decision-making in prefrontal cortex, insula, thalamus, and other regions. IGT alternative metrics have neural correlates consistent with decision-making theory as those difference scores reduce heterogeneity in cognitive processes. The CD-AB metric construct failure may reflect an artificial amalgamation of processes. The D-A metric appears to more successfully combine multiple levels of representation (dorsolateral prefrontal cortex, sub-cortical, cerebellar). (JINS, 2012, 18, 585–594)

The purpose of this study was to compare the neuropsychological outcome from blast-related versus non-blast related mild traumatic brain injury (MTBI). Participants were 56 U.S. military service members who sustained an MTBI, divided into two groups based on mechanism of injury: (a) non-blast related (Non-blast; n = 21), and (b) blast plus secondary blunt trauma (Blast Plus; n = 35). All participants had sustained their injury in theatre whilst deployed during Operation Iraqi Freedom or Operation Enduring Freedom. Patients had been seen for neuropsychological evaluation at Walter Reed Army Medical Center on average 4.4 months (SD = 4.1) post-injury. Measures included 14 clinical scales from the Personality Assessment Inventory (PAI) and 12 common neurocognitive measures. For the PAI, there were no significant differences between groups on all scales (p > .05). However, medium effect sizes were found for the Depression (d = .49) and Stress (d = .47) scales (i.e., Blast Plus > Non-blast). On the neurocognitive measures, after controlling for the influence of psychological distress (i.e., Depression, Stress), there were no differences between the Non-blast and Blast Plus groups on all measures. These findings provide little evidence to suggest that blast exposure plus secondary blunt trauma results in worse cognitive or psychological recovery than blunt trauma alone. (JINS, 2012, 18, 595–605)

Serial assessments are commonplace in neuropsychological practice and used to document cognitive trajectory for many clinical conditions. However, true change scores may be distorted by measurement error, repeated exposure to the assessment instrument, or person variables. The present study provides reliable change indices (RCI) for the Boston Naming Test, derived from a sample of 844 cognitively normal adults aged 56 years and older. All participants were retested between 9 and 24 months after their baseline exam. Results showed that a 4-point decline during a 9–15 month retest period or a 6-point decline during a 16–24 month retest period represents reliable change. These cutoff values were further characterized as a function of a person's age and family history of dementia. These findings may help clinicians and researchers to characterize with greater precision the temporal changes in confrontation naming ability. (JINS, 2012, 18, 375–378)

The ability to appraise one's own ability has been found to have an important role in the recovery and quality of life of clinical populations. Examinee and task variables have been found to influence metacognition in healthy students; however the effect of these variables on the metacognitive accuracy of adults with neurological insult, such as traumatic brain injury (TBI), remains unknown. Twenty-two adults with moderate and severe TBI and a matched sample of healthy adults participated in this study examining the influence of item sequencing on metacognitive functioning. Retrospective confidence judgments were collected while participants completed a modified version of the Matrix Reasoning subtest. Significant influence of item sequence order was found, revealing better metacognitive abilities and performance when participants completed tasks where item difficulty progressed in order from easy to difficult. We interpret these findings to suggest that the sequencing of item difficulty offers “anchors” for gauging and adjusting to task demands. (JINS, 2012, 18, 379–383)

The purpose of this study was to investigate the relative effects of unilateral (left-sided) versus bilateral deep brain stimulation (DBS) of the subthalamic nucleus (STN) on verbal fluency. To do this, 10 Parkinson's disease patients with predominantly bilateral motor symptoms who received bilateral STN DBS were compared with 6 patients suffering from predominantly unilateral symptoms who received STN DBS on the left side only. The results suggest that unilateral STN DBS of the speech dominant hemisphere is associated with significantly less declines in measures of verbal fluency as compared to bilateral stimulation. (JINS, 2012, 18, 606–611)