Siyuan Liu

Neural Correlates of Lyrical Improvisation: An fMRI Study of Freestyle Rap

The neural correlates of creativity are poorly understood. Freestyle rap provides a unique opportunity to study spontaneous lyrical improvisation, a multidimensional form of creativity at the interface of music and language. Here we use functional magnetic resonance imaging to characterize this process. Task contrast analyses indicate that improvised performance is characterized by dissociated activity in medial and dorsolateral prefrontal cortices, providing a context in which stimulus-independent behaviors may unfold in the absence of conscious monitoring and volitional control. Connectivity analyses reveal widespread improvisation-related correlations between medial prefrontal, cingulate motor, perisylvian cortices and amygdala, suggesting the emergence of a network linking motivation, language, affect and movement. Lyrical improvisation appears to be characterized by altered relationships between regions coupling intention and action, in which conventional executive control may be bypassed and motor control directed by cingulate motor mechanisms. These functional reorganizations may facilitate the initial improvisatory phase of creative behavior.

Brain activity and connectivity during poetry composition: Toward a multidimensional model of the creative process

Creativity, a multifaceted construct, can be studied in various ways, for example, investigating phases of the creative process, quality of the creative product, or the impact of expertise. Previous neuroimaging studies have assessed these individually. Believing that each of these interacting features must be examined simultaneously to develop a comprehensive understanding of creative behavior, we examined poetry composition, assessing process, product, and expertise in a single experiment. Distinct activation patterns were associated with generation and revision, two major phases of the creative process. Medial prefrontal cortex (MPFC) was active during both phases, yet responses in dorsolateral prefrontal and parietal executive systems (DLPFC/IPS) were phase‐dependent, indicating that while motivation remains unchanged, cognitive control is attenuated during generation and re‐engaged during revision. Experts showed significantly stronger deactivation of DLPFC/IPS during generation, suggesting that they may more effectively suspend cognitive control. Importantly however, similar overall patterns were observed in both groups, indicating the same cognitive resources are available to experts and novices alike. Quality of poetry, assessed by an independent panel, was associated with divergent connectivity patterns in experts and novices, centered upon MPFC (for technical facility) and DLPFC/IPS (for innovation), suggesting a mechanism by which experts produce higher quality poetry. Crucially, each of these three key features can be understood in the context of a single neurocognitive model characterized by dynamic interactions between medial prefrontal areas regulating motivation, dorsolateral prefrontal, and parietal areas regulating cognitive control and the association of these regions with language, sensorimotor, limbic, and subcortical areas distributed throughout the brain. Hum Brain Mapp 36:3351–3372, 2015.

Neural correlates and network connectivity underlying narrative production and comprehension: A combined fMRI and PET study

The neural correlates of narrative production and comprehension remain poorly understood. Here, using positron emission tomography (PET), functional magnetic resonance imaging (fMRI), contrast and functional network connectivity analyses we comprehensively characterize the neural mechanisms underlying these complex behaviors. Eighteen healthy subjects told and listened to fictional stories during scanning. In addition to traditional language areas (e.g., left inferior frontal and posterior middle temporal gyri), both narrative production and comprehension engaged regions associated with mentalizing and situation model construction (e.g., dorsomedial prefrontal cortex, precuneus and inferior parietal lobules) as well as neocortical premotor areas, such as the pre-supplementary motor area and left dorsal premotor cortex. Narrative comprehension alone showed marked bilaterality, activating right hemisphere homologs of perisylvian language areas. Narrative production remained predominantly left lateralized, uniquely activating executive and motor-related regions essential to language formulation and articulation. Connectivity analyses revealed strong associations between language areas and the superior and middle temporal gyri during both tasks. However, only during storytelling were these same language-related regions connected to cortical and subcortical motor regions. In contrast, during story comprehension alone, they were strongly linked to regions supporting mentalizing. Thus, when employed in a more complex, ecologically-valid context, language production and comprehension show both overlapping and idiosyncratic patterns of activation and functional connectivity. Importantly, in each case the language system is integrated with regions that support other cognitive and sensorimotor domains.

Embodied comprehension of stories: Interactions between language regions and modality-specific neural systems

The embodied view of language processing proposes that comprehension involves multimodal simulations, a process that retrieves a comprehenders perceptual, motor, and affective knowledge through reactivation of the neural systems responsible for perception, action, and emotion. Although evidence in support of this idea is growing, the contemporary neuroanatomical model of language suggests that comprehension largely emerges as a result of interactions between frontotemporal language areas in the left hemisphere. If modality-specific neural systems are involved in comprehension, they are not likely to operate in isolation but should interact with the brain regions critical to language processing. However, little is known about the ways in which language and modality-specific neural systems interact. To investigate this issue, we conducted a functional MRI study in which participants listened to stories that contained visually vivid, action-based, and emotionally charged content. Activity of neural systems associated with visual-spatial, motor, and affective processing were selectively modulated by the relevant story content. Importantly, when functional connectivity patterns associated with the left inferior frontal gyrus (LIFG), the left posterior middle temporal gyrus (pMTG), and the bilateral anterior temporal lobes (aTL) were compared, both LIFG and pMTG, but not the aTL, showed enhanced connectivity with the three modality-specific systems relevant to the story content. Taken together, our results suggest that language regions are engaged in perceptual, motor, and affective simulations of the described situation, which manifest through their interactions with modality-specific systems. On the basis of our results and past research, we propose that the LIFG and pMTG play unique roles in multimodal simulations during story comprehension.

Denoising the speaking brain: Toward a robust technique for correcting artifact-contaminated fMRI data under severe motion

A comprehensive set of methods based on spatial independent component analysis (sICA) is presented as a robust technique for artifact removal, applicable to a broad range of functional magnetic resonance imaging (fMRI) experiments that have been plagued by motion-related artifacts. Although the applications of sICA for fMRI denoising have been studied previously, three fundamental elements of this approach have not been established as follows: 1) a mechanistically-based ground truth for component classification; 2) a general framework for evaluating the performance and generalizability of automated classifiers; and 3) a reliable method for validating the effectiveness of denoising. Here we perform a thorough investigation of these issues and demonstrate the power of our technique by resolving the problem of severe imaging artifacts associated with continuous overt speech production. As a key methodological feature, a dual-mask sICA method is proposed to isolate a variety of imaging artifacts by directly revealing their extracerebral spatial origins. It also plays an important role for understanding the mechanistic properties of noise components in conjunction with temporal measures of physical or physiological motion. The potentials of a spatially-based machine learning classifier and the general criteria for feature selection have both been examined, in order to maximize the performance and generalizability of automated component classification. The effectiveness of denoising is quantitatively validated by comparing the activation maps of fMRI with those of positron emission tomography acquired under the same task conditions. The general applicability of this technique is further demonstrated by the successful reduction of distance-dependent effect of head motion on resting-state functional connectivity.

Lesion analysis of language production deficits in aphasia

Background: Three aspects of language production are impaired to different degrees in individuals with post-stroke aphasia: ability to repeat words and nonwords, name pictures, and produce sentences. These impairments often persist into the chronic stages, and the neuroanatomical distribution of lesions associated with chronicity of each of these impairments is incompletely understood. Aims: The primary objective of this study was to investigate the lesion correlates of picture naming, sentence production, and nonword repetition deficits in the same participant group because most prior lesion studies have mapped single language impairments. The broader goal of this study was to investigate the extent and degree of overlap and uniqueness among lesions resulting in these deficits in order to advance the current understanding of functional subdivision of neuroanatomical regions involved in language production. Methods & Procedures: In this study, lesion-symptom mapping was used to determine if specific cortical regions are associated with nonword repetition, picture naming, and sentence production scores. Structural brain images and behavioural performance of 31 individuals with post-stroke left hemisphere lesions and a diagnosis of aphasia were used in the lesion analysis. Outcomes & Results: Each impairment was associated with mostly unique, but a few shared lesions. Overall, sentence and repetition deficits were associated with left anterior perisylvian lesions, including the pars opercularis and triangularis of the inferior frontal lobe, anterior superior temporal gyrus, anterior portions of the supramarginal gyrus, the putamen, and anterior portions of the insula. In contrast, impaired picture naming was associated with posterior perisylvian lesions including major portions of the inferior parietal lobe and middle temporal gyrus. The distribution of lesions in the insula was consistent with this antero-posterior perisylvian gradient. Significant voxels in the posterior planum temporale were associated with a combination of all three deficits. Conclusions: These findings emphasise the participation of each perisylvian region in multiple linguistic functions, suggesting a many(functions)-to-many(networks) framework while also identifying functional subdivisions within each region.

The correspondence problem: which brain maps are significantly similar?

A critical issue in many neuroimaging studies is the comparison between brain maps. How should we test the hypothesis that two or more brain maps are partially convergent or overlap to a significant extent? This “correspondence problem” affects, for example, the interpretation of comparisons between task-based patterns of functional activation, resting-state networks or modules, and neuroanatomical landmarks. In published work, this problem has been addressed with remarkable variability in terms of methodological approaches and statistical rigor. In this paper, we address the correspondence problem using a spatial permutation framework to generate null models of overlap, by applying random rotations to spherical representations of the cortical surface. We use this approach to derive clusters of cognitive functions that are significantly similar in terms of their functional neuroatomical substrates. In addition, using publicly available data, we formally demonstrate the correspondence between maps of task-based functional activity, resting-state fMRI networks and gyral-based anatomical landmarks. We provide open-access code to implement the methods presented for two commonly-used tools for surface based cortical analysis. This spatial permutation approach constitutes a useful advance over widely-used methods for the comparison of cortical maps, and thereby opens up new possibilities for the integration of diverse neuroimaging data.

Normative brain size variation and brain shape diversity in humans

Shifts in brain regions with brain size Brain size among normal humans varies as much as twofold. Reardon et al. surveyed the cortical and subcortical structure of more than 3000 human brains by noninvasive imaging (see the Perspective by Van Essen). They found that the scaling of different regions across the range of brain sizes is not consistent: Some brain regions are metabolically costly and are favored in larger brains. This shifts the balance between associative and sensorimotor brain systems in a brain size–dependent way. Science, this issue p. 1222; see also p. 1184 A metabolically expensive brain network is preferentially expanded in individuals that have larger brains. Brain size variation over primate evolution and human development is associated with shifts in the proportions of different brain regions. Individual brain size can vary almost twofold among typically developing humans, but the consequences of this for brain organization remain poorly understood. Using in vivo neuroimaging data from more than 3000 individuals, we find that larger human brains show greater areal expansion in distributed frontoparietal cortical networks and related subcortical regions than in limbic, sensory, and motor systems. This areal redistribution recapitulates cortical remodeling across evolution, manifests by early childhood in humans, and is linked to multiple markers of heightened metabolic cost and neuronal connectivity. Thus, human brain shape is systematically coupled to naturally occurring variations in brain size through a scaling map that integrates spatiotemporally diverse aspects of neurobiology.

On testing for spatial correspondence between maps of human brain structure and function

Abstract A critical issue in many neuroimaging studies is the comparison between brain maps. Nonetheless, it remains unclear how one should test hypotheses focused on the overlap or spatial correspondence between two or more brain maps. This “correspondence problem” affects, for example, the interpretation of comparisons between task‐based patterns of functional activation, resting‐state networks or modules, and neuroanatomical landmarks. To date, this problem has been addressed with remarkable variability in terms of methodological approaches and statistical rigor. In this paper, we address the correspondence problem using a spatial permutation framework to generate null models of overlap by applying random rotations to spherical representations of the cortical surface, an approach for which we also provide a theoretical statistical foundation. We use this method to derive clusters of cognitive functions that are correlated in terms of their functional neuroatomical substrates. In addition, using publicly available data, we formally demonstrate the correspondence between maps of task‐based functional activity, resting‐state fMRI networks and gyral‐based anatomical landmarks. We provide open‐access code to implement the methods presented for two commonly‐used tools for surface based cortical analysis (https://www.github.com/spin‐test). This spatial permutation approach constitutes a useful advance over widely‐used methods for the comparison of cortical maps, thereby opening new possibilities for the integration of diverse neuroimaging data. HighlightsA new method is developed to test the anatomical correspondence between brain maps.Random rotational permutations generate rigorous null models of correspondence.The correspondence of structural, functional and resting‐state maps is quantified.These methods are publicly available for future applications.

Overt social interaction and resting state in autism: core and contextual neural features

Conversation is an important and ubiquitous social behavior. Individuals with Autism Spectrum Disorder (autism) without intellectual disability often have normal language abilities but deficits in social aspects of verbal interaction like pragmatics, prosody, and eye contact. Previous studies of resting state activity suggest that intrinsic connections among neural circuits involved with social processing are disrupted in autism, but to date no neuroimaging study has examined neural activity during the most commonplace yet challenging social task: spontaneous conversation. Here we used functional MRI to scan autistic males (N=19) without intellectual disability and age- and IQ-matched typically developing controls (N=20) while they engaged in a total of 193 face-to-face interactions. Participants completed two kinds of tasks: Conversation, which had high social demand, and Repetition, which had low social demand. Autistic individuals showed abnormally increased task-driven inter-regional temporal correlation relative to controls, especially among social processing regions and during high social demand. Furthermore, these increased correlations were associated with more severe autism symptoms. These results were then compared with previously-acquired resting-state data (56 Autism, 62 Control participants). While some inter-regional correlation levels varied by task or rest context, others were strikingly similar across both task and rest, namely increased correlation among the thalamus, dorsal and ventral striatum, somatomotor, temporal and prefrontal cortex in the autistic individuals, relative to the control groups. These results suggest a basic distinction. Autistic cortico-cortical interactions vary by context, tending to increase relative to controls during Task and decrease during Rest. In contrast, striato- and thalamocortical relationships with socially engaged brain regions are increased in both Task and Rest, and may be core to the condition of autism. Abbreviations BOLD Blood Oxygen Level Dependent IFG inferior frontal gyrus IQ Intelligence Quotient PCA principal component analysis ROI region of interest SD standard deviation STS superior temporal sulcus TR Repetition Time