Predicting Alcohol-Related Memory Problems in Older Adults: A Machine Learning Study with Multi-Domain Features
Chella KamarajanAshwini K. PandeyDavid B. ChorlianJacquelyn L. MeyersSivan KinreichGayathri PandeyStacey Saenz de ViteriJian ZhangWeipeng KuangPeter B. BarrFazil AlıevAndrey P. AnokhinMartin H. PlaweckiSamuel KupermanLaura AlmasyAlison MerikangasSarah J. BrislinLance O. BauerVictor HesselbrockGrace ChanJohn R. KramerDongbing LaiSarah M. HartzLaura J. BierutVivia V. McCutcheonKathleen K. BucholzDanielle M. DickMarc A. SchuckitHoward J. EdenbergBernice Porjesz
0
Citation
0
Reference
10
Related Paper
Abstract:
Memory problems are common among older adults with a history of alcohol use disorder (AUD). Employing a machine learning framework, the current study investigates the use of multi-domain features to classify individuals with and without alcohol-induced memory problems. A group of 94 individuals (ages 50-81 years) with alcohol-induced memory problems (Memory group) were compared with a matched Control group who did not have memory problems. The Random Forests model identified specific features from each domain that contributed to the classification of Memory vs. Control group (AUC=88.29%). Specifically, individuals from the Memory group manifested a predominant pattern of hyperconnectivity across the default mode network regions except some connections involving anterior cingulate cortex which were predominantly hypoconnected. Other significant contributing features were (i) polygenic risk scores for AUD, (ii) alcohol consumption and related health consequences during the past 5 years, such as health problems, past negative experiences, withdrawal symptoms, and the largest number of drinks in a day during the past 12 months, and (iii) elevated neuroticism and increased harm avoidance, and fewer positive “uplift” life events. At the neural systems level, hyperconnectivity across the default mode network regions, including the connections across the hippocampal hub regions, in individuals with memory problems may indicate dysregulation in neural information processing. Overall, the study outlines the importance of utilizing multidomain features, consisting of resting-state brain connectivity collected ~18 years ago, together with personality, life experiences, polygenic risk, and alcohol consumption and related consequences, to predict alcohol-related memory problems that arise in later life.Keywords:
Alcohol use disorder
Chronic fatigue syndrome (CFS) is characterized by a debilitating fatigue of unknown aetiology. Patients who suffer from CFS report a variety of physical complaints as well as neuropsychological complaints. Therefore, it is conceivable that the CNS plays a role in the pathophysiology of CFS. The purpose of this study was to investigate neural correlates of CFS, and specifically whether there exists a linkage between disturbances in the motor system and CFS. We measured behavioural performance and cerebral activity using rapid event-related functional MRI in 16 CFS patients and 16 matched healthy controls while they were engaged in a motor imagery task and a control visual imagery task. CFS patients were considerably slower on performance of both tasks, but the increase in reaction time with increasing task load was similar between the groups. Both groups used largely overlapping neural resources. However, during the motor imagery task, CFS patients evoked stronger responses in visually related structures. Furthermore, there was a marked between-groups difference during erroneous performance. In both groups, dorsal anterior cingulate cortex was specifically activated during error trials. Conversely, ventral anterior cingulate cortex was active when healthy controls made an error, but remained inactive when CFS patients made an error. Our results support the notion that CFS may be associated with dysfunctional motor planning. Furthermore, the between-groups differences observed during erroneous performance point to motivational disturbances as a crucial component of CFS.
Dysfunctional family
Error-related negativity
Cite
Citations (143)
Harm avoidance is one of the four temperament dimensions in Cloninger's psychobiological model of personality (Cloninger et al. , 1993). In this model, personality is conceptualized as having four temperament dimensions (novelty seeking, harm avoidance, reward dependence, persistence) and three character dimensions (self-directedness, cooperativeness, self-transcendence). Individuals high in harm avoidance (HA) tend to be described as “worrying and pessimistic; fearful and doubtful; shy; and fatigable.” Those with low HA scores are “relaxed and optimistic; bold and confident; outgoing; and vigorous.” HA was initially proposed as a personality trait that is independent of the state of depression. However, other reports have found HA to be positively correlated with the severity of depression (Hansenne et al ., 1999; Hirano et al ., 2002). Repeated within-subject measures have been used to explore the stability of HA in major depression. However, conflicting results are reported in the literature and research in HA is largely with adults (Chien and Dunner, 1996; Marijnissen et al ., 2002). The aim of this study is to investigate whether the personality trait of HA is stable over a 12-month period in a group of older people with a history of depression. This is a first report on the stability of HA in an exclusive elderly sample.
Reward dependence
Cooperativeness
Persistence (discontinuity)
Trait
Depression
Cite
Citations (13)
Alcohol use disorder
Cooperativeness
Reward dependence
Alcohol Dependence
Persistence (discontinuity)
Cite
Citations (3)
The default mode,as an intrinsic system of the brain,plays an very important role in self-retrospection,cognition and episodic memory retrieval processes.Many chronic brain diseases may effect changes in the default mode.Epileptic seizure can at once evoke abnormal brain activation and impact the default mode network.The authors present an overview of the features of the default mode network as well as the relationship of epilepsy with negative activation and the default mode network.
Cite
Citations (0)
Abstract Because corticotrophin‐releasing hormone (CRH) plays a central role in stress regulation, the possible role of CRH1 polymorphism for anxiety‐related personality variables such as harm avoidance possibly associated with alcoholism was studied. The research instruments used to phenotype patients were adopted partly from the US collaborative study of the genetics of alcoholism and include a number of personality inventories such as the temperament and character inventory (TCI). Based on the examination of 170 alcoholic subjects no association was found between CRH1 receptor haplotypes of four single nuclotid polymorphisms (SNPs) and low and high temperament traits of harm avoidance, novelty seeking and reward dependence. The possible implications of these findings are discussed.
Alcohol Dependence
Association (psychology)
Reward dependence
Cite
Citations (16)
Cued speech
Cingulate cortex
Neural Activity
Cite
Citations (22)
Alcohol use disorder (AUD) has been related to aberrant functional connectivity (FC) in the salience network (SN), executive control network (ECN), and default mode network (DMN). However, there is a lack of comprehensive and simultaneous examination of these networks in patients with AUD and of their relation to potential anatomical changes. We aimed to comprehensively examine the alteration in FC in the three networks in AUD patients, and the correlation of the alteration with anatomical/structural changes (volume) in the neural areas implicated in these networks, by applying voxel-based morphometry (VBM) and region of interest-to-region of interest connectivity analysis simultaneously. In all, 22 patients with AUD and 22 healthy adults participated in the study and underwent T1 magnetic resonance imaging. Patients with AUD showed increased FCs within the DMN and SN networks, especially in terms of connectivity of the frontal areas and bilateral hippocampi. They also showed decreased FCs in the ECN. In addition, there was significant volume reduction in these areas (frontal areas and hippocampus). The increased FCs within the frontal areas or bilateral hippocampi showed a negative correlation with gray matter volume of these areas in AUD patients. Our findings add to the empirical evidence that the frontal lobe and hippocampi are critical areas that are vulnerable to functional and structural changes due to AUD.
Alcohol use disorder
Salience (neuroscience)
Frontal lobe
Voxel-based morphometry
Cite
Citations (17)
Task-positive network
Posterior cingulate
Cite
Citations (0)
Orbitofrontal cortex
Alcohol use disorder
Cingulate cortex
Cite
Citations (4)
The importance of personality characteristics in the diagnosis and treatment of gambling disorder (GD) and alcohol use disorder (AUD) is often highlighted in scientific literature. This study aimed to test predictions about the associations of temperament and character in chronic AUD patients with comorbid GD symptoms and without them.Chronic AUD patients enrolled from an inpatient clinic were divided in two groups based on cluster analysis, AUD patients with (AUD + GD group: n = 30) and without (AUD group: n = 68) GD symptoms. Severity of GD symptoms and personality dimensions (Cloninger's Temperament and Character Inventory Revised, TCI-R) were assessed. Associations of tested variables were analysed with analysis of covariance, one-sample and independent sample t-tests.GD symptoms proved to be a clustering factor in terms of personality, where AUD + GD group expressed a more maladaptive personality profile. Compared to Hungarian normative TCI-R scores, both patient groups showed elevated levels of Harm Avoidance and Novelty Seeking with lower scores of Self-directedness, while the AUD + GD group scored lower on Persistence and Cooperation as well. The AUD + GD group reported significantly higher levels of Harm Avoidance, with lower scores of Reward Dependence compared to the AUD group.Comorbid GD symptom severity is an important factor in chronic AUD, where AUD patients with comorbid GD symptoms exhibited a more maladaptive personality constellation than singular AUD patients. These emphasize the need of special attention for comorbid GD symptoms in AUD, since treatment recommendations and prognosis for them may also differ.
Alcohol use disorder
Personality profile
Alcohol Dependence
Cite
Citations (6)