Face recognition ability is manifest in early dynamic decoding of face-orientation selectivity-Evidence from multi-variate pattern analysis of the neural response.

Although humans are considered to be face experts, there is a well-established reliable variation in the degree to which neurotypical individuals are able to learn and recognise faces. While many behavioural studies have characterised these differences, studies that seek to relate the neuronal response to standardised behavioural measures of ability remain relatively scarce, particularly so for the time-resolved approaches and the early response to face stimuli. In the present study we make use of a relatively recent methodological advance, multi-variate pattern analysis (MVPA), to decode the time course of the neural response to faces compared to other object categories (inverted faces, objects). Importantly, for the first time, we directly relate metrics of this decoding assessed at the individual level to gold-standard measures of behavioural face processing ability assessed in an independent task. Thirty-nine participants completed the behavioural Cambridge Face Memory Test (CFMT), then viewed images of faces and houses (presented upright and inverted) while their neural activity was measured via electroencephalography. Significant decoding of both face orientation and face category were observed in all individual participants. Decoding of face orientation, a marker of more advanced face processing, was earlier and stronger in participants with higher levels of face expertise, while decoding of face category information was earlier but not stronger for individuals with greater face expertise. Taken together these results provide a marker of significant differences in the early neuronal response to faces from around 100 ms post stimulus as a function of behavioural expertise with faces.

The time course of creativity: Multivariate classification of default and executive network contributions to creative cognition over time.

Research indicates that creative cognition depends on both associative and controlled processes, corresponding to the brain's default mode network (DMN) and executive control network (ECN) networks. However, outstanding questions include how the DMN and ECN operate over time during creative task performance, and whether creative cognition involves distinct generative and evaluative stages. To address these questions, we used multivariate pattern analysis (MVPA) to assess how the DMN and ECN contribute to creative cognition over three successive time phases during the production of a single creative idea. Training classifiers to predict trial condition (creative vs non-creative), we used classification accuracy as a measure of the extent of creative activity in each brain network and time phase. Across both networks, classification accuracy was highest in early phases, decreased in mid phases, and increased again in later phases, following a U-shaped curve. Notably, classification accuracy was significantly greater in the ECN than the DMN during early phases, while differences between networks at later time phases were non-significant. We also computed correlations between classification accuracy and human-rated creative performance, to assess how relevant the creative activity in each network was to the creative quality of ideas. In line with expectations, classification accuracy in the DMN was most related to creative quality in early phases, decreasing in later phases, while classification accuracy in the ECN was least related to creative quality in early phases, increasing in later phases. Given the theorized roles of the DMN in generation and the ECN in evaluation, we interpret these results as tentative evidence for the existence of separate generative and evaluative stages in creative cognition that depend on distinct neural substrates.

Identification of radiographic characteristics associated with pain in hallux valgus patients: A preliminary machine learning study.

Objective: To investigate the association between the structural deformity and foot pain in hallux valgus (HV) patients using a multi-variate pattern analysis (MVPA) approach. Methods: Plain radiographic metrics were calculated from 36 painful and 36 pain-free HV feet. In analysis 1, univariate analyses were performed to investigate the clinical and radiographic differences between painful and pain-free HV. In analysis 2, we investigated the pattern differences for radiographic metrics between these two groups using a MVPA approach utilizing a support vector machine. In analysis 3, sequential backward selection and exhaustive search were performed as a feature-selection procedure to identify an optimal feature subtype. In analysis 4, hierarchical clustering analysis was used to identify the optimal radiographic HV subtype associated with pain in HV. Results: We found that: (1) relative to feet with pain-free HV, the painful ones exhibited a higher hallux valgus angle, i.e., the magnitude of distal metatarsal and phalangeal deviation; (2) painful HV could be accurately differentiated from pain-free HV via MVPA. Using sequential backward selection and exhaustive search, a 5-feature subset was identified with optimal performance for classifying HV as either painful or pain-free; and (3) by applying hierarchical clustering analysis, a radiographic subtype with an 80% pain incidence was identified. Conclusion: The pain in HV is multifactorial and associated with a radiographic pattern measured by various angles on plain radiographs. The combination of hallux valgus angle, inter-phalangeal angle, distal metatarsal articular angle, metatarsal cuneiform angle and metatarsal protrusion distance showed the optimal classification performance between painful and pain-free HV.

Testing the distributed representation hypothesis in object recognition in two open datasets.

Neural representation has long been thought to follow the modularity hypothesis, which states that each type of information corresponds to a specific brain area. Though supported by many studies, this hypothesis surfers the pitfall of inefficiency for information encoding. To overcome difficulties the modularity representation hypothesis faced, researchers have proposed that information may be distributed represented in a specific brain area. The distributed representation hypothesis along with the multi-variate pattern approaches have made great success in detecting representation patterns in the previous decade. However, this hypothesis implicitly requires that the pattern should be transformed in a consistent way with respect to all of the represented information in the specific brain area. And the accuracy and validity of this prediction have never been thoroughly tested. Here in the present study, we tested this prediction in two open datasets compiling the object recognition. We validated the distributed representation patterns in the lateral occipital complex/ventral temporal gyrus where all six classifiers were capable of predicting the correct category represented. Furthermore, we correlated the classifiers' decision function values to the bold signals and found that the decision function value of the logistic regression classifier was exclusively correlated with activities of the same brain area in both datasets. These results support the distributed representation hypothesis and suggest that our neural system may be embedded within the algorithm of a specific classifier.

Neural signatures of vigilance decrements predict behavioural errors before they occur.

There are many monitoring environments, such as railway control, in which lapses of attention can have tragic consequences. Problematically, sustained monitoring for rare targets is difficult, with more misses and longer reaction times over time. What changes in the brain underpin these 'vigilance decrements'? We designed a multiple-object monitoring (MOM) paradigm to examine how the neural representation of information varied with target frequency and time performing the task. Behavioural performance decreased over time for the rare target (monitoring) condition, but not for a frequent target (active) condition. This was mirrored in neural decoding using magnetoencephalography: coding of critical information declined more during monitoring versus active conditions along the experiment. We developed new analyses that can predict behavioural errors from the neural data more than a second before they occurred. This facilitates pre-empting behavioural errors due to lapses in attention and provides new insight into the neural correlates of vigilance decrements.

Temporal embedding and spatiotemporal feature selection boost multi-voxel pattern analysis decoding accuracy.

BACKGROUND: In fMRI decoding, temporal embedding of spatial features of the brain allows the incorporation of brain activity dynamics into the multivariate pattern classification process, and provides enriched information about stimulus-specific response patterns and potentially improved prediction accuracy. NEW METHOD: This study investigates the possibility of enhancing the classification performance by exploring temporal embedding, to identify the optimum combination of spatiotemporal features based on their classification performance. We investigated the importance of spatiotemporal feature selection using a slow event-related design adapted from the classic Haxby study (Haxby et al., 2001). Data were collected using a multiband fMRI sequence with temporal resolution of 0.568 s. COMPARISON WITH EXISTING METHODS: A wide range of spatiotemporal observations were created as various combinations of spatiotemporal features. Using both random forest, and support vector machine, classifiers prediction accuracies for these combinations were then compared with the single spatial multivariate pattern approach that uses only a single temporal observation. RESULTS: Our findings showed that, on average, spatiotemporal feature selection improved prediction accuracy. Moreover, the random forest algorithm outperformed the support vector machine and benefitted from temporal information to a greater extent. CONCLUSIONS: As expected, the most influential temporal durations were found to be around the peak of the hemodynamic response function, a few seconds after the stimuli onset until -4 s after the peak of the hemodynamic response function. The superiority of spatiotemporal feature selection over single time-point spatial approaches invites future work to design optimal approaches that incorporate spatiotemporal dependencies into feature selection for decoding.

Characterizing the neural signature of face processing in Williams syndrome via multivariate pattern analysis and event related potentials.

Face recognition ability is often reported to be a relative strength in Williams syndrome (WS). Yet methodological issues associated with the supporting research, and evidence that atypical face processing mechanisms may drive outcomes 'in the typical range', challenge these simplistic characterisations of this important social ability. Detailed investigations of face processing abilities in WS both at a behavioural and neural level provide critical insights. Here, we behaviourally characterised face recognition ability in 18 individuals with WS comparatively to typically developing children and adult control groups. A subset of 11 participants with WS as well as chronologically age matched typical adults further took part in an EEG task where they were asked to attentively view a series of upright and inverted faces and houses. State-of-the-art multivariate pattern analysis (MVPA) was used alongside standard ERP analysis to obtain a detailed characterisation of the neural profile associated with 1) viewing faces as an overall category (by examining neural activity associated with upright faces and houses), and to 2) the canonical upright configuration of a face, critically associated with expertise in typical development and often linked with holistic processing (upright and inverted faces). Our results show that while face recognition ability is not on average at a chronological age-appropriate level in individuals with WS, it nonetheless appears to be a relative strength within their cognitive profile. Furthermore, all participants with WS revealed a differential pattern of neural activity to faces compared to objects, showing a distinct response to faces as a category, as well as a differential neural pattern for upright vs. inverted faces. Nonetheless, an atypical profile of face orientation classification was found in WS, suggesting that this group differs from typical individuals in their face processing mechanisms. Through this innovative application of MVPA, alongside the high temporal resolution of EEG, we provide important new insights into the neural processing of faces in WS.

Decoding the dynamic representation of facial expressions of emotion in explicit and incidental tasks.

Faces transmit a wealth of important social signals. While previous studies have elucidated the network of cortical regions important for perception of facial expression, and the associated temporal components such as the P100, N170 and EPN, it is still unclear how task constraints may shape the representation of facial expression (or other face categories) in these networks. In the present experiment, we used Multivariate Pattern Analysis (MVPA) with EEG to investigate the neural information available across time about two important face categories (expression and identity) when those categories are either perceived under explicit (e.g. decoding facial expression category from the EEG when task is on expression) or incidental task contexts (e.g. decoding facial expression category from the EEG when task is on identity). Decoding of both face categories, across both task contexts, peaked in time-windows spanning 91-170 ms (across posterior electrodes). Peak decoding of expression, however, was not affected by task context whereas peak decoding of identity was significantly reduced under incidental processing conditions. In addition, errors in EEG decoding correlated with errors in behavioral categorization under explicit processing for both expression and identity, however under incidental conditions only errors in EEG decoding of expression correlated with behavior. Furthermore, decoding time-courses and the spatial pattern of informative electrodes showed consistently better decoding of identity under explicit conditions at later-time periods, with weak evidence for similar effects for decoding of expression at isolated time-windows. Taken together, these results reveal differences and commonalities in the processing of face categories under explicit Vs incidental task contexts and suggest that facial expressions are processed to a richer degree under incidental processing conditions, consistent with prior work indicating the relative automaticity by which emotion is processed. Our work further demonstrates the utility in applying multivariate decoding analyses to EEG for revealing the dynamics of face perception.

CoSMoMVPA: Multi-Modal Multivariate Pattern Analysis of Neuroimaging Data in Matlab/GNU Octave.

Recent years have seen an increase in the popularity of multivariate pattern (MVP) analysis of functional magnetic resonance (fMRI) data, and, to a much lesser extent, magneto- and electro-encephalography (M/EEG) data. We present CoSMoMVPA, a lightweight MVPA (MVP analysis) toolbox implemented in the intersection of the Matlab and GNU Octave languages, that treats both fMRI and M/EEG data as first-class citizens. CoSMoMVPA supports all state-of-the-art MVP analysis techniques, including searchlight analyses, classification, correlations, representational similarity analysis, and the time generalization method. These can be used to address both data-driven and hypothesis-driven questions about neural organization and representations, both within and across: space, time, frequency bands, neuroimaging modalities, individuals, and species. It uses a uniform data representation of fMRI data in the volume or on the surface, and of M/EEG data at the sensor and source level. Through various external toolboxes, it directly supports reading and writing a variety of fMRI and M/EEG neuroimaging formats, and, where applicable, can convert between them. As a result, it can be integrated readily in existing pipelines and used with existing preprocessed datasets. CoSMoMVPA overloads the traditional volumetric searchlight concept to support neighborhoods for M/EEG and surface-based fMRI data, which supports localization of multivariate effects of interest across space, time, and frequency dimensions. CoSMoMVPA also provides a generalized approach to multiple comparison correction across these dimensions using Threshold-Free Cluster Enhancement with state-of-the-art clustering and permutation techniques. CoSMoMVPA is highly modular and uses abstractions to provide a uniform interface for a variety of MVP measures. Typical analyses require a few lines of code, making it accessible to beginner users. At the same time, expert programmers can easily extend its functionality. CoSMoMVPA comes with extensive documentation, including a variety of runnable demonstration scripts and analysis exercises (with example data and solutions). It uses best software engineering practices including version control, distributed development, an automated test suite, and continuous integration testing. It can be used with the proprietary Matlab and the free GNU Octave software, and it complies with open source distribution platforms such as NeuroDebian. CoSMoMVPA is Free/Open Source Software under the permissive MIT license. Website: http://cosmomvpa.org Source code: https://github.com/CoSMoMVPA/CoSMoMVPA.

Mapping causal functional contributions derived from the clinical assessment of brain damage after stroke.

Lesion analysis reveals causal contributions of brain regions to mental functions, aiding the understanding of normal brain function as well as rehabilitation of brain-damaged patients. We applied a novel lesion inference technique based on game theory, Multi-perturbation Shapley value Analysis (MSA), to a large clinical lesion dataset. We used MSA to analyze the lesion patterns of 148 acute stroke patients together with their neurological deficits, as assessed by the National Institutes of Health Stroke Scale (NIHSS). The results revealed regional functional contributions to essential behavioral and cognitive functions as reflected in the NIHSS, particularly by subcortical structures. There were also side specific differences of functional contributions between the right and left hemispheric brain regions which may reflect the dominance of the left hemispheric syndrome aphasia in the NIHSS. Comparison of MSA to established lesion inference methods demonstrated the feasibility of the approach for analyzing clinical data and indicated its capability for objectively inferring functional contributions from multiple injured, potentially interacting sites, at the cost of having to predict the outcome of unknown lesion configurations. The analysis of regional functional contributions to neurological symptoms measured by the NIHSS contributes to the interpretation of this widely used standardized stroke scale in clinical practice as well as clinical trials and provides a first approximation of a 'map of stroke'.

Individual identification research of amnestic mild cognitive impairment based on support vector machine / 医学研究生学报

Objective In recent years , multivariate pattern analysis ( MVPA) method was proposed and considered to be a promising tool for automated identification of various neuropsychiatric populations .Support vector machine ( SVM) is one of the most widely used methods of MVPA .Using SVM classifier for MVPA of amnestic mild cognitive impairment (aMCI) and normal control (NC) group, the present study aims to build an individual diagnostic model with significant discriminative power and investigate the gray matter abnor-malities of aMCI patients . Methods Fifty-one aMCI patients and 68 normal controls were scanned on the 3-Tesla magnetic resonance imaging (MRI) for high-resolution T1-weighted images.Gray matter volume map was calculated for each subject and used as features for subsequent discriminative analysis .We first applied feature selection to remove redundant information and reduce feature dimension , and then trained an SVM classifier . Leave-one-out cross validation ( LOOCV) was used to estimate the performance of the classifier , and finally the most discriminative features were identified . Results The proposed classifier achieved a classification accuracy of 83.19%with a sensitivity of 76.47%and a specificity of 88.24%.In ad-dition, the area under the receiver operating characteristic (ROC) curve was 0.8368.Further analysis revealed that the most discrimi-native features for classification included bilateral parahippocampal gyri , bilateral hippocampi , bilateral amygdala , bilateral thalamus , right cingulate , right precuneus , left caudate , left superior temporal gyrus , left middle temporal gyrus , left insula and left orbitofrontal cortex. Conclusion The proposed classification model has achieved significant accuracy for aMCI prediction , and it also displayed the whole brain gray matter atrophy pattern in aMCI patients .It suggests that the proposed method may have important implications for early clinical diagnosis of aMCI patients .