Internal representation of hierarchical sequences involves the default network

Self-similarity and recursion as default modes in human cognition

Humans generate recursive hierarchies in a variety of domains, including linguistic, social and visuo-spatial modalities. The ability to represent recursive structures has been hypothesized to increase the efficiency of hierarchical processing. Theoretical work together with recent empirical findings suggests that the ability to represent the self-similar structure of hierarchical recursive stimuli may be supported by internal neural representations that compress raw external information and increase efficiency. In order to explicitly test whether the representation of recursive hierarchies depends on internalized rules we compared the processing of visual hierarchies represented either as recursive or non-recursive, using task-free resting-state fMRI data. We aimed to evaluate the relationship between task-evoked functional networks induced by cognitive representations with the corresponding resting-state architecture. We observed increased connectivity within Default Mode Network (DMN) related brain areas during the representation of recursion, while non-recursive representations yielded increased connectivity within the Fronto-Parietal Control-Network. Our results suggest that human hierarchical information processing using recursion is supported by the DMN. In particular, the representation of recursion seems to constitute an internally-biased mode of information-processing that is mediated by both the core and dorsal-medial subsystems of the DMN. Compressed internal rule representations mediated by the DMN may help humans to represent and process hierarchical structures in complex environments by considerably reducing information processing load.

Noise concerns and post-processing procedures in cerebral blood flow (CBF) and cerebral blood volume (CBV) functional magnetic resonance imaging

Functional neuroimaging with blood oxygenation level-dependent (BOLD) contrast has emerged as the most popular method for evaluating qualitative changes in brain function in humans. At typical human field strengths (1.5-3.0T), BOLD contrast provides a measure of changes in transverse water relaxation rates in and around capillary and venous blood, and as such provides only a surrogate marker of brain function that depends on dynamic changes in hemodynamics (e.g., cerebral blood flow and volume) and metabolism (e.g., oxygen extraction fraction and the cerebral metabolic rate of oxygen consumption). Alternative functional neuroimaging methods that are specifically sensitive to these constituents of the BOLD signal are being developed and applied in a growing number of clinical and neuroscience applications of quantitative cerebral physiology. These methods require additional considerations for interpreting and quantifying their contrast responsibly. Here, an overview of two popular methods, arterial spin labeling and vascular space occupancy, is presented specifically in the context of functional neuroimaging. Appropriate post-processing and experimental acquisition strategies are summarized with the motivation of reducing sensitivity to noise and unintended signal sources and improving quantitative accuracy of cerebral hemodynamics.

Connectivity in the human brain dissociates entropy and complexity of auditory inputs

Complex systems are described according to two central dimensions: (a) the randomness of their output, quantified via entropy; and (b) their complexity, which reflects the organization of a system's generators. Whereas some approaches hold that complexity can be reduced to uncertainty or entropy, an axiom of complexity science is that signals with very high or very low entropy are generated by relatively non-complex systems, while complex systems typically generate outputs with entropy peaking between these two extremes. In understanding their environment, individuals would benefit from coding for both input entropy and complexity; entropy indexes uncertainty and can inform probabilistic coding strategies, whereas complexity reflects a concise and abstract representation of the underlying environmental configuration, which can serve independent purposes, e.g., as a template for generalization and rapid comparisons between environments. Using functional neuroimaging, we demonstrate that, in response to passively processed auditory inputs, functional integration patterns in the human brain track both the entropy and complexity of the auditory signal. Connectivity between several brain regions scaled monotonically with input entropy, suggesting sensitivity to uncertainty, whereas connectivity between other regions tracked entropy in a convex manner consistent with sensitivity to input complexity. These findings suggest that the human brain simultaneously tracks the uncertainty of sensory data and effectively models their environmental generators.

Goal-congruent default network activity facilitates cognitive control

Substantial neuroimaging evidence suggests that spontaneous engagement of the default network impairs performance on tasks requiring executive control. We investigated whether this impairment depends on the congruence between executive control demands and internal mentation. We hypothesized that activation of the default network might enhance performance on an executive control task if control processes engage long-term memory representations that are supported by the default network. Using fMRI, we scanned 36 healthy young adult humans on a novel two-back task requiring working memory for famous and anonymous faces. In this task, participants (1) matched anonymous faces interleaved with anonymous face, (2) matched anonymous faces interleaved with a famous face, or (3) matched a famous faces interleaved with an anonymous face. As predicted, we observed a facilitation effect when matching famous faces, compared with anonymous faces. We also observed greater activation of the default network during these famous face-matching trials. The results suggest that activation of the default network can contribute to task performance during an externally directed executive control task. Our findings provide evidence that successful activation of the default network in a contextually relevant manner facilitates goal-directed cognition.

Potential use and challenges of functional connectivity mapping in intractable epilepsy

This review focuses on the use of resting-state functional magnetic resonance imaging data to assess functional connectivity in the human brain and its application in intractable epilepsy. This approach has the potential to predict outcomes for a given surgical procedure based on the pre-surgical functional organization of the brain. Functional connectivity can also identify cortical regions that are organized differently in epilepsy patients either as a direct function of the disease or through indirect compensatory responses. Functional connectivity mapping may help identify epileptogenic tissue, whether this is a single focal location or a network of seizure-generating tissues. This review covers the basics of connectivity analysis and discusses particular issues associated with analyzing such data. These issues include how to define nodes, as well as differences between connectivity analyses of individual nodes, groups of nodes, and whole-brain assessment at the voxel level. The need for arbitrary thresholds in some connectivity analyses is discussed and a solution to this problem is reviewed. Overall, functional connectivity analysis is becoming an important tool for assessing functional brain organization in epilepsy.

The intrinsic connectivity distribution: a novel contrast measure reflecting voxel level functional connectivity

Resting-state fMRI (rs-fMRI) holds promise as a clinical tool to characterize and monitor the phenotype of different neurological and psychiatric disorders. The most common analysis approach requires the definition of one or more regions-of-interest (ROIs). However this need for a priori ROI information makes rs-fMRI inadequate to survey functional connectivity differences associated with a range of neurological disorders where the ROI information may not be available. A second problem encountered in fMRI measures of connectivity is the need for an arbitrary correlation threshold to determine whether or not two areas are connected. This is problematic because in many cases the differences in tissue connectivity between disease groups and/or control subjects are threshold dependent. In this work we propose a novel voxel-based contrast mechanism for rs-fMRI, the intrinsic connectivity distribution (ICD), that neither requires a priori information to define a ROI, nor an arbitrary threshold to define a connection. We show the sensitivity of previous methods to the choice of connection thresholds and evaluate ICD using a survey study comparing young adults born prematurely to healthy term control subjects. Functional connectivity differences were found in hypothesized language processing areas in the left temporal-parietal areas. In addition, significant clinically-relevant differences were found between preterm and term control subjects, highlighting the importance of whole brain surveys independent of a priori information.

Multiple sensitivity profiles to diversity and transition structure in non-stationary input

Recent formalizations suggest that the human brain codes for the degree of order in the environment and utilizes this knowledge to optimize perception and performance in the immediate future. However, the neural bases of how the brain spontaneously codes for order are poorly understood. It has been shown that activity in lateral temporal cortex and the hippocampus is linearly correlated with the order of short visual series under tasks requiring attention to the input and when series order is invariant over time. Here, we examined if sensitivity to order is manifested in both linear and non-linear BOLD response profiles, quantified the degree to which order-sensitive regions operate as a functional network, and evaluated these questions using a paradigm in which performance of the ongoing task could be completed without any attention to the stimulus whose order was manipulated. Participants listened to a 10-minute sequence of tones characterized by non-stationary order, and fMRI identified cortical regions sensitive to time-varying statistical features of this input. Activity in perisylvian regions was negatively correlated with input diversity, quantified via Shannon's Entropy. Activity in ventral premotor, lateral temporal, and insular regions was correlated linearly, parabolically, or via a step-function with the strength of transition constraints in the series, quantified via Markov Entropy. Granger-causality analysis revealed that order-sensitive regions form a functional network, with regions showing non-linear responses to order associated with more afferent connectivity than those showing linear responses. These findings identify networks that spontaneously code and respond to diverse aspects of order via multiple response profiles, and that play a central role in generating and gating predictive neural activity.