Cerebral Cortex

Unethical actions and decisions may distort human memory in two aspects: memory accuracy and metacognition. However, the neural and computational mechanisms underlying the metacognition distortion caused by repeated dishonesty remain largely unknown. Here, we performed two fMRI studies, including one replication study, with an information-sending task in the scanner. The main moral decision task in the scanner involves consistency and reward as two main factors, combined with a pre-scan and post-scan memory test together with mouse tracking. With multiple dimensions of metrics to measure metacognition, we test whether the inter-subject metacognition change correlates with how participants trade off consistency and reward. We find that the compression of representational geometry of reward in the orbitofrontal cortex (OFC) is correlated with both immediate and delayed metacognition changes. Also, the functional connectivity between the dorsolateral prefrontal cortex (DLPFC) and the left temporoparietal junction (lTPJ) under dishonest responses can predict both immediate and delayed metacognition changes in memory. These results suggest that decision-making, emotion, and memory-related brain regions together play a key role in metacognition change after immoral action, shedding light on the neural mechanism of the complex interplay between moral decisions, cognitive processes, and memory distortion. I did that, says my memory. I could not have done that, says my pride, and remains inexorable. Eventually - the memory yields. --Nietzsche[1]

In humans, many neurobiological features of the cortex--including gene expression patterns, microstructure, and functional connectivity--vary systematically along a sensorimotor-association (S-A) axis of brain organisation. To date, it is still poorly understood whether inter-individual differences in patterns of S-A axis capture these robust spatial relationships across neurobiological properties observed at the group-level. Here, we examine inter-individual differences in structural and functional properties of the S-A axis, namely cortical microstructure, geodesic distances, and the functional gradient, in a sample of young adults from the Human Connectome Project (N = 992, including 328 twins). We quantified heritable variation associated with inter-individual differences in the S-A axis, and assessed whether structural and functional properties that are highly spatially correlated at the group-level also share genetic underpinnings. To consider measurement errors in resting-state functional connectivity data and their impact on properties of the S-A axis, we used a multivariate twin design capable of disentangling individual-level variation in both intra- and inter-individual differences. After accounting for some of the intra-individual variation, we found average heritable individual differences in both the functional gradient (htwin2 = 57%), cortical microstructure (htwin2 = 43%), and geodesic distances (htwin2 = 34%). However, these genetic influences were mostly distinct and deviated from group-level patterns. In particular, we found no significant genetic correlation between the functional gradient and microstructure, while we found both positive and negative genetic associations between the functional gradient and geodesic distances. Our approach highlights the complexity of genetic contributions to brain organisation and may have potential implications for understanding cognitive variability within the S-A axis framework.

As long anticipated (Sandberg and Bostrom 2008; Seung 2012; Szigeti et al. 2014), connectomics is providing a new foundation for brain simulation by replacing theoretical assumptions about network connectivity with solid empirical facts. Connectomics also yields detailed information about neuronal morphology, which is useful for simulating the biophysics of single neurons (Yang et al. 2016; Meier and Borst 2019; T. X. Liu et al. 2022). Here I introduce a formalism for simulating a brain as a network of synapses interacting via an effective resistance matrix. By computing this matrix for fly visual interneurons, I find evidence that some neurons may be true multi-output devices, neither well approximated as "point neurons" (Lappalainen et al. 2024; Shiu et al. 2024), nor as collections of functionally independent compartments (Meier and Borst 2019). Within a linear approximation, such a neuron is instead equivalent to a hierarchy of virtual neurons that spatially pool over multiple length scales. The computational powers of multi-output neurons may support highly sophisticated normalizations in the fly visual system (Seung 2024a).

Recently, a neuroscientific approach has revealed that humans understand language while subconsciously predicting the next word from the preceding context. Most studies on human word prediction have investigated the correlations between brain activity while reading or listening to sentences on functional magnetic resonance imaging (fMRI) and the predictive difficulty of each word in a sentence calculated by the N-gram language model. However, because of its low temporal resolution, fMRI is not optimal for identifying the changes in brain activity that accompany language comprehension. In addition, the N-gram language model is a simple computational structure that does not account for the structure of the human brain. Furthermore, it is necessary for humans to retain information prior to the N-1 word in order to form a contextual understanding of a presented story. Therefore, in the present study, we measured brain activity using magnetoencephalography (MEG), which has a higher temporal resolution than fMRI, and calculated the prediction difficulty of words using a long short-term memory language model (LSTMLM), which is based on a neural network inspired by the structure of the human brain and has longer information retention than the N-gram language model. We then identified the brain regions involved in language prediction during Japanese-language speech listening using encoding and decoding analyses. In addition to surprisal-related regions revealed in previous studies, such as the superior temporal gyrus, fusiform gyrus, and temporal pole, we also found relationships between surprisal and brain activity in other regions, including the insula, superior temporal sulcus, and middle temporal gyrus, which are believed to be involved in longer-term, sentence-level cognitive processing.

This study examines whether bilingual exposure has a profound effect on the functional organization of the developing human brain during infancy. Recent behavioural research attests that monolingual vs. bilingual experience affects cognitive and linguistic processes already during the first months of life. However, to what extent the intrinsic organization of the infant human brain adapts to monolingual vs. bilingual environments is unclear. We measured spontaneous hemodynamic brain activity using functional near-infrared spectroscopy (fNIRS) in a large cohort (N=99) of 4-month-old monolingual and bilingual infants. We implemented well-established analysis approaches of functional brain imaging that enabled us to reveal the functional organization of the infant brain in large-scale cortical networks, and to perform group-level comparisons (i.e., monolingual vs. bilingual groups) in a reliable manner. Our results revealed no differences between the intrinsic functional organization of the developing monolingual and bilingual infant brain at 4 months of age.

The functional significance of the two prominent language-related ERP components N400 and P600 is still under debate. It has recently been suggested that one important dimension along which the two vary is in terms of automaticity versus attentional control, with N400 amplitudes reflecting more automatic and P600 amplitudes reflecting more controlled aspects of sentence comprehension. The availability of executive resources necessary for controlled processes depends on sustained attention, which fluctuates over time. Here, we thus tested whether P600 and N400 amplitudes depend on the level of sustained attention. We re-analyzed EEG and behavioral data from a sentence processing task by Sassenhagen & Bornkessel-Schlesewsky (2015, Cortex), which included sentences with morphosyntactic and semantic violations. Participants read sentences phrase by phrase and indicated whether a sentence contained any type of anomaly as soon as they had the relevant information. To quantify the varying degree of sustained attention, we extracted a moving reaction time coefficient of variation over the entire course of the task. We found that the P600 amplitude was significantly larger during periods of low reaction time variability (high sustained attention) than in periods of high reaction time variability (low sustained attention). In contrast, the amplitude of the N400 was not affected by reaction time variability. These results thus suggest that the P600 component is sensitive to sustained attention while the N400 component is not, which provides independent evidence for accounts suggesting that P600 amplitudes reflect more controlled and N400 amplitudes more automatic aspects of sentence comprehension.

This neuroimaging study investigated the neural infrastructure of sentence-level language production. We compared brain activation patterns, as measured with BOLD-fMRI, during production of sentences which differed in verb argument structures (intransitives, transitives, ditransitives) and the lexical status of the verb (known verbs or pseudo-verbs). An example for the type of sentence to be produced started a mini-block of six sentences with the same structure. For each trial, participants were first given the (pseudo-)verb followed by three geometric shapes to serve as verb arguments in the sentences. Production of sentences with known verbs yielded greater activation compared to those with pseudo-verbs in the core language network of left inferior frontal gyrus, the left posterior middle temporal gyrus, and a more posterior middle temporal region extending into the angular gyrus (LpMTG/AG), analogous to effects observed in language comprehension. Increasing the number of verb arguments led to greater activation in an overlapping left pMTG/AG area, particularly for known verbs, as well as in the bilateral precuneus. Thus, producing sentences with more complex structures using existing verbs lead to increased activation in the language network, suggesting some reliance on memory retrieval of stored lexical-syntactic information during sentence production. This study thus provides evidence from sentence-level language production in line with functional models of the language network that have so far been mainly based on single word production, comprehension and processing in aphasia.

Empty categories are unpronounced elements with syntactic properties that play a central role in theories of sentence structure. Although there are several types within these categories, the neural basis for distinguishing among them remains unclear. Using magnetoencephalography, we investigate whether the brain distinguishes between two Japanese sentence structures: Control and raising. Although these constructions appear similar on the surface, they are argued to involve different types of empty categories. In theoretical analyses, the control-type empty category, which is called PRO, is often treated as an anaphoric element, similar to reflexives such as himself and herself, whereas the raising-type empty category is a noun phrase trace. Twenty-six native Japanese speakers participated in a reading task under three experimental conditions: Control, raising, and baseline. Source estimates were computed, and condition differences were tested using spatiotemporal cluster-based permutation t-tests. We observed late left-hemispheric differences at approximately 700-800 ms after the critical verb. The control condition elicited larger responses than the raising condition, with activity centered in the temporal cortex spanning the middle temporal gyrus and the superior temporal sulcus and gyrus and extending into the anterior insula and the supramarginal gyrus. In addition, the control condition elicited larger late responses than the baseline condition in a broader left fronto-temporal distribution, including the inferior frontal cortex, anterior temporal cortex, and insula. These results provide source-level evidence that brain activity in the left language network differs between the control and raising conditions in Japanese during online sentence comprehension. Furthermore, they suggest that we can distinguish empty category types in the brain.

The role of cortical oscillations in brain function has been extensively debated, resulting in a variety of theoretical frameworks. Using inter-leaved simultaneous EEG-fMRI, we examined the layer-specific relationship between oscillatory activity and visual processing. We could demonstrate that{gamma} -band activity positively correlates with feature specific signals in superficial layers, but we were able to report a deep layer contribution as well. In addition, we could demonstrate that -band power not only correlates negatively with the feature unspecific BOLD signal, but related to feature specific BOLD as well. Lower frequency was pre-dominantly related to feature unspecific superficial layer BOLD, while upper frequency was found to be related to feature specific BOLD in superficial and deep layers. We conclude that the role of -band oscillations extends beyond widespread inhibition and might be involved in active stimulus processing to the level of visual features.

A fundamental property of the neocortex is its columnar organization in many species. Generally, neurons of the same column share stimulus preferences and have strong anatomical connections across layers. These features suggest that neurons within a column operate as one unified network. Other features, like the different patterns of input and output connections of neurons located in separate layers and systematic differences in feature tuning, hint at a more segregated and possibly flexible functional organization of neurons within a column. To distinguish between these views of columnar processing, we conducted laminar recordings in macaques area V1 while they performed a demanding attention task. We found three separate regions with strong gamma oscillatory current source density (CSD) signals, one each in the supragranular, granular, and infragranular laminar domains. Their characteristics differed significantly in terms of their dominant gamma frequency and attention-dependent modulation of their gramma power and gamma frequency. In line, spiking activity in the supragranular, infragranular, and upper part of the granular domain exhibited strong phase coherence with their domains CSD signals but showed much weaker coherence with the other domains CSD signals. These results indicate that columnar processing involves a certain degree of independence between neurons in the three laminar domains, consistent with the assumption of multiple, separate intracolumnar ensembles. Such a functional organization offers various possibilities for dynamic network configuration, indicating that neurons in a column are not restricted to operate as one unified network. Thus, the findings open interesting new possibilities for future concepts and investigations on flexible, dynamic cortical ensemble formation and selective information processing.

In childhood and adolescence, functional brain networks go through different stages of development, and the levels of connectivity within and between these networks change with age. The developmental trajectory of these large-scale networks of the brain have been extensively investigated; however, many aspects of our social brain and its developmental patterns remain unclear. This study employed a cross-sectional design to investigate the brains of 753 children and adolescents (ages 5-15) while they watched a movie. This research investigates the functional distinctness and developmental synchronicity of brain areas implicated in social cognition, such as empathy and affective and cognitive theory of mind, across childhood and adolescence using generalized additive models and inter-region group analysis. Our findings suggest that social cognition components networks such as cognitive and affective theory of mind and empathy exhibit distinct developmental trajectories throughout childhood and adolescence. The findings support the theory that social cognitive networks are developmentally distinct from each other, even in the absence of task-specific paradigms.

One organizing principle of the human brain is hemispheric specialization, or the dominance of a specific function or cognitive process in one hemisphere or the other. Previously, Wang et al. (2014) identified networks putatively associated with language and attention as being specialized to the left and right hemispheres, respectively; and a dual-specialization of the executive control network. However, it remains unknown which networks are specialized when specialization is examined within individuals using a higher resolution parcellation, as well as which connections are contributing the most to a given networks specialization. In the present study, we estimated network specialization across three datasets using the autonomy index and a novel method of deconstructing network specialization. After examining the reliability of these methods as implemented on an individual level, we addressed two hypotheses. First, we hypothesized that the most specialized networks would include those associated with language, visuospatial attention, and executive control. Second, we hypothesized that within-network contributions to specialization would follow a within-between network gradient or a specialization gradient. We found that the majority of networks exhibited greater within-hemisphere connectivity than between-hemisphere connectivity. Among the most specialized networks were networks associated with language, attention, and executive control. Additionally, we found that the greatest network contributions were within-network, followed by those from specialized networks. Significance StatementHemispheric specialization is a characteristic of brain organization that describes when a function draws on one hemisphere of the brain more than the other. We sought to identify the most specialized brain networks within individuals, as well as which connections contribute the most to a given networks specialization. Among the most specialized networks were those associated with language, attention, and executive control. Unexpectedly, we also identified networks associated with emotion/memory and theory of mind as highly specialized. Additionally, we found support for guiding principles of brain organization generally, such that within-network connections contributed most to a given networks specialization followed by connections from other specialized networks. These results have implications for identifying potential variations of network contributions in individuals with neurodevelopmental conditions.

The early development of the prefrontal cortex is crucial for higher cognitive functions. However, current research presents inconsistent findings regarding whether intra-prefrontal connectivity increases or decreases in infants younger than six months. Do dynamic changes in connection strength across different states over time carry information about prefrontal maturation? This study used functional near-infrared spectroscopy (fNIRS) to record prefrontal brain activity in 48 healthy infants aged 1-8 months during natural sleep and auditory stimulation. By analyzing the fluctuations in frequency-domain characteristics of functional connectivity (FC) and various brain network properties, we found that: under auditory stimulation, the intensity of FC fluctuations in the ultra-low frequency range was positively correlated with age; while in the resting state, the fluctuation intensity of network properties in relatively higher frequency bands decreased with age. Furthermore, auditory stimulation reconfigured the energy distribution of network fluctuations, shifting it towards higher frequency bands. These results suggest that the early development of the infant prefrontal internal network is characterized by state-dependent optimization of its dynamic fluctuation properties, shedding light on the developmental tuning of functional network dynamics in infancy.

With advancing age, the distinctiveness of neural representations of information declines. While the finding of this so-called age-related neural dedifferentiation in category-selective neural regions is well-described, how neural dedifferentiation manifests at the level of large-scale functional networks is less understood. Furthermore, the relationship between age-related changes in network organization and dedifferentiation is unknown. Here, we investigated age-related neural dedifferentiation of category-selective regions as well as whole-brain functional networks. We additionally examined age differences in connectivity of category-selective regions to the rest of the brain. Younger and older adults viewed blocks of face and house stimuli while performing memory encoding and retrieval in the fMRI scanner. We found an age-related decline in neural distinctiveness for faces in the fusiform gyrus (FG) and for houses in the parahippocampal gyrus (PHG). Functional connectivity analyses revealed age-related dedifferentiation of global network structure as well as age differences in the connectivity profiles to category-selective regions. Together, our findings suggest that age-related neural dedifferentiation manifests both in regional categorical representations as well as in whole- brain functional networks. HighlightsO_LICategory representations are less distinctive, or dedifferentiated, in older adults C_LIO_LIFunctional networks are less segregated in older adults C_LIO_LIOlder adults reveal less connectivity between fusiform gyrus and visual cortices C_LI

Training to improve working memory is associated with changes in prefrontal activation and confers lasting benefits, some of which generalize to untrained tasks, though the issue remains contentious and the neural substrate underlying such transfer are unknown. To assess how neural activity changes induced by training transfer across tasks, we recorded single units, multi-unit activity (MUA) and local field potentials (LFP) with chronic electrode arrays implanted in the prefrontal cortex of two monkeys, as they were trained to perform cognitive tasks. Mastering different tasks was associated with distinct changes in neural activity, which included redistribution of power across frequency bands in the LFP, recruitment of larger numbers of MUA sites, and increase or decrease of mean neural activity across single units. In every training phase, changes induced by the actively learned task transferred to an untrained control task, which remained the same across the training period. The results explicate the neural basis through which training can transfer across cognitive tasks.

Auditory exposure plays crucial roles in shaping healthy brain development and generating lateralization of functional network organization. However, little is known about whether and how an initial lack of auditory exposure in early infancy may disrupt development of functional network lateralization. We addressed this issue by recruiting 55 infants with congenital sensorineural hearing loss (SNHL) and 60 typically developing (TD) controls. Resting-state fNIRS imaging data were acquired to construct hemispheric cerebral networks, and graph theory was applied to quantify the topological characteristics of hemispheric networks. The infants with SNHL exhibited efficient small-world characteristic within each hemispheric network, however, the lateralization of functional network efficiency was substantially disrupted. Compared with TD infants with significantly increased network efficiency lateralized toward left hemisphere with age, the SNHL infants did not exhibit the emergence and development of such cerebral lateralization. Furthermore, the increased leftward asymmetry in nodal efficiency with age was found in TD but not in SNHL infants. Interestingly, the degree of hearing loss had no correlation with lateralization strength in the SNHL group. These results suggest that SNHL infants exhibited disrupted development of cortical lateralization in functional network organization, and highlight the importance of auditory stimulation-promoted multisensory functional integration in early infancy.

The question of whether repeated studies bring more variability or less to our brain is a critical problem before scientist continue their further study about memory. In the past ten years, a series of neural pattern representation studies have found that under the condition of spaced learning, the neural pattern similarity(NPS) between two encoding stages increases, and the researchers claimed that these results support the idea that repeated studies bring more similarity along with repetition in our brain, which is conflicted with the encoding variability theory(Feng et al., 2019; Y. Lu et al., 2015; Xue et al., 2010, 2011a). However, we doubt this viewpoint because we think the difference between encoding processing cannot be used to represent the difference between memory states. In current experiments, we used a new experimental paradigm with a longer lag and elaboration learning task to test the encoding variability theory. By comparing the difference between neural pattern dissimilarity(NPDS1)(spaced learning - one-time learning) and NPDS2(massed learning - one-time learning) in the final test (retrieval) stage, we get the result that the NPDS1 was significantly greater than NPDS2 in the parietal lobe of 400ms and the right frontal lobe of 600ms, which is more fitting to the encoding variability mechanism. However, we believe that there is no contradiction between these two experimental evidences. On the contrary, we think they reflect different aspects of the process of spaced learning. We propose that the deficient processing in encoding stage exactly lead to less encoding variability in our memory. More importantly, this result gives us reason to double the paper published ten years ago in Science, which claimed repetition brings greater neural pattern similarity.

Attending to different features of a scene can alter the responses of neurons in early- and mid- level visual areas but the nature of this change depends on both the (top down) attentional task and the (bottom up) visual stimulus. One outstanding question is the spatial scale at which cortex is modulated by attention to low-level stimulus features such as shape, contrast and orientation. It is unclear whether the recruitment of neurons to particular tasks occurs at an area level or at the level of intra-areal sub-populations, or whether the critical factor is a change in the way that areas communicate with each other. Here we use functional magnetic resonance imaging (fMRI) and psychophysics, to ask how areas known to be involved in processing different visual features (orientation, contrast and shape) are modulated as participants switch between tasks based on those features while the visual stimulus itself is effectively constant. At a univariate level, we find almost no feature-specific bottom-up or top-down responses in the areas we examine. However, multivariate analyses reveal a complex pattern of voxel-level modulation driven by attentional task. Connectivity analyses also demonstrate flexible and selective patterns of connectivity between early visual areas as a function of attentional focus. Overall, we find that attention alters the sensitivity and connectivity of neuronal subpopulations within individual early visual areas but, surprisingly, not the univariate response amplitudes of the areas themselves.

The striatum is postulated to play a role in gating cortical processing during goal-oriented behavior. However, the underlying circuit structure for striatal gating remains unclear. Deviating from previous approaches which typically treat the striatum as a homogenous structure or small compartments, we took a functional connectivity approach that utilizes the entire anatomical space of the caudate nucleus and examined its functional relationship with the cortex and how that relationship changes with age. We defined the topography of the caudate functional connectivity with the rest of the brain using three publicly available resting-state fMRI data samples. There were several key findings. First, our results revealed two stable gradients of connectivity patterns across the caudate: medial-lateral (M-L) and anterior-posterior (A-P) axes, which supports findings in previous anatomical studies of non-human primates that there is more than one organizational principle. Second, the differential connectivity patterns along the caudates M-L gradient were not limited to single structures but rather organized with respect to large-scale neural networks; in particular, networks associated with internal orienting behavior are closely linked to the medial extent of the caudate whereas networks associated with external orienting behavior are closely linked to the lateral extent of the caudate. Third, we found a decrease in the integrity of M-L organization with healthy aging which was associated with age-related changes in behavioral measures of flexible control. In sum, the caudate shows a topographic organization with respect to large-scale networks in the human brain and changes this organization seem to have implications for age-related decline in flexible control of behavior.

Emotional prosody processing involves a widespread network of brain regions, but the specific roles of the cerebellum and basal ganglia in explicit and implicit tasks are not well known or understood. This study investigated how the cerebellum and basal ganglia contribute to explicit (emotion categorization) and implicit (gender categorization) processing of emotional prosody, namely when attention is directly versus implicitly oriented towards the emotion of the voice stimuli, respectively. Twenty-eight healthy French-speaking participants (average age: 65 years old) underwent high-resolution functional MRI while performing explicit and implicit vocal emotion processing tasks. Neuroimaging results revealed--and replicated--that both tasks recruited a widespread network, including the superior temporal cortex, inferior frontal cortex, primary motor and somatosensory cortices, basal ganglia, and cerebellum. The explicit task elicited stronger activations in the basal ganglia (caudate nucleus, putamen) and cerebellar regions (Crus I/II, lobules VI, VIIb, and X), consistent with higher cognitive control demands. In contrast, the implicit task was associated with activations in cerebellar lobules IV-V, VI, VIII, and IX, along with the thalamus. Regression-based functional connectivity analyses further demonstrated stronger connectivity between the right cerebellar lobule IX and the putamen, as well as the cerebellar vermis (XII), particularly during implicit processing. These findings highlight the distinct contributions of the cerebellum and basal ganglia to emotional prosody processing, with explicit tasks engaging associative and cognitive control networks, while implicit tasks rely more on sensorimotor and automatic neural processing mechanisms.