Isolated anterior cerebral artery occlusion: an atypical form of moyamoya disease.

BACKGROUND: The relationship between anterior cerebral artery (ACA) occlusion and moyamoya disease (MMD) has rarely been studied. In this study, we focused on a special type of MMD: isolated ACA-occlusive MMD. We investigated clinical attributes, genotypes and progression risk factors in patients with ACA-occlusive MMD, providing initial insights into the relationship between ACA occlusion and MMD. METHODS: We retrospectively analysed digital subtraction angiography (DSA) from 2486 patients and diagnosed 139 patients with ACA-occlusive MMD. RNF213 p.R4810K (rs112735431) mutation analysis was performed. Patients were categorised into progression and non-progression groups based on whether they progressed to typical MMD. Differences in clinical characteristics, neuropsychological assessment, radiological findings and genotypes were evaluated. Logistic regression analyses identified risk factors for ACA-occlusive MMD progression. RESULTS: The median age of patients with ACA-occlusive MMD was 36 years, and the primary symptom was transient ischaemic attack (TIA). 72.3% of ACA-occlusive MMD patients had cognitive decline. Of 116 patients who underwent RNF213 gene mutation analysis, 90 patients (77.6%) carried the RNF213 p.R4810K GG allele and 26 (22.4%) carried the GA allele. Of 102 patients with follow-up DSA data, 40 patients (39.2%) progressed. Kaplan-Meier curve estimates indicated a higher incidence of ischaemic stroke in the progression group during follow-up (p=0.035). Younger age (p=0.041), RNF213 p.R4810K GA genotype (p=0.037) and poor collateral compensation from the middle cerebral artery (MCA) to ACA (p<0.001) were risk factors of ACA-occlusive MMD progression to typical MMD. CONCLUSIONS: Cognitive decline and TIA might be the main manifestations of ACA-occlusive MMD. Isolated ACA occlusion may be an early signal of MMD. The initial lesion site of MMD is not strictly confined to the terminal portion of the internal carotid artery. Younger patients, patients with RNF213 p.R4810K GA genotype or those with inadequate MCA-to-ACA compensation are more likely to develop typical MMD.

Adaptive mechanisms facilitate robust performance in noise and in reverberation in an auditory categorization model.

For robust vocalization perception, the auditory system must generalize over variability in vocalization production as well as variability arising from the listening environment (e.g., noise and reverberation). We previously demonstrated using guinea pig and marmoset vocalizations that a hierarchical model generalized over production variability by detecting sparse intermediate-complexity features that are maximally informative about vocalization category from a dense spectrotemporal input representation. Here, we explore three biologically feasible model extensions to generalize over environmental variability: (1) training in degraded conditions, (2) adaptation to sound statistics in the spectrotemporal stage and (3) sensitivity adjustment at the feature detection stage. All mechanisms improved vocalization categorization performance, but improvement trends varied across degradation type and vocalization type. One or more adaptive mechanisms were required for model performance to approach the behavioral performance of guinea pigs on a vocalization categorization task. These results highlight the contributions of adaptive mechanisms at multiple auditory processing stages to achieve robust auditory categorization.

Optimal features for auditory categorization.

Humans and vocal animals use vocalizations to communicate with members of their species. A necessary function of auditory perception is to generalize across the high variability inherent in vocalization production and classify them into behaviorally distinct categories ('words' or 'call types'). Here, we demonstrate that detecting mid-level features in calls achieves production-invariant classification. Starting from randomly chosen marmoset call features, we use a greedy search algorithm to determine the most informative and least redundant features necessary for call classification. High classification performance is achieved using only 10-20 features per call type. Predictions of tuning properties of putative feature-selective neurons accurately match some observed auditory cortical responses. This feature-based approach also succeeds for call categorization in other species, and for other complex classification tasks such as caller identification. Our results suggest that high-level neural representations of sounds are based on task-dependent features optimized for specific computational goals.