The neural code for face orientation in the human fusiform face area

Emergence of brain-like mirror-symmetric viewpoint tuning in convolutional neural networks

Primates can recognize objects despite 3D geometric variations such as in-depth rotations. The computational mechanisms that give rise to such invariances are yet to be fully understood. A curious case of partial invariance occurs in the macaque face-patch AL and in fully connected layers of deep convolutional networks in which neurons respond similarly to mirror-symmetric views (e.g. left and right profiles). Why does this tuning develop? Here, we propose a simple learning-driven explanation for mirror-symmetric viewpoint tuning. We show that mirror-symmetric viewpoint tuning for faces emerges in the fully connected layers of convolutional deep neural networks trained on object recognition tasks, even when the training dataset does not include faces. First, using 3D objects rendered from multiple views as test stimuli, we demonstrate that mirror-symmetric viewpoint tuning in convolutional neural network models is not unique to faces: it emerges for multiple object categories with bilateral symmetry. Second, we show why this invariance emerges in the models. Learning to discriminate among bilaterally symmetric object categories induces reflection-equivariant intermediate representations. AL-like mirror-symmetric tuning is achieved when such equivariant responses are spatially pooled by downstream units with sufficiently large receptive fields. These results explain how mirror-symmetric viewpoint tuning can emerge in neural networks, providing a theory of how they might emerge in the primate brain. Our theory predicts that mirror-symmetric viewpoint tuning can emerge as a consequence of exposure to bilaterally symmetric objects beyond the category of faces, and that it can generalize beyond previously experienced object categories.

A Unifying Model for Discordant and Concordant Results in Human Neuroimaging Studies of Facial Viewpoint Selectivity

Recognizing faces regardless of their viewpoint is critical for social interactions. Traditional theories hold that view-selective early visual representations gradually become tolerant to viewpoint changes along the ventral visual hierarchy. Newer theories, based on single-neuron monkey electrophysiological recordings, suggest a three-stage architecture including an intermediate face-selective patch abruptly achieving invariance to mirror-symmetric face views. Human studies combining neuroimaging and multivariate pattern analysis (MVPA) have provided convergent evidence of view selectivity in early visual areas. However, contradictory conclusions have been reached concerning the existence in humans of a mirror-symmetric representation like that observed in macaques. We believe these contradictions arise from low-level stimulus confounds and data analysis choices. To probe for low-level confounds, we analyzed images from two face databases. Analyses of image luminance and contrast revealed biases across face views described by even polynomials-i.e., mirror-symmetric. To explain major trends across neuroimaging studies, we constructed a network model incorporating three constraints: cortical magnification, convergent feedforward projections, and interhemispheric connections. Given the identified low-level biases, we show that a gradual increase of interhemispheric connections across network-layers is sufficient to replicate view-tuning in early processing stages and mirror-symmetry in later stages. Data analysis decisions-pattern dissimilarity measure and data recentering-accounted for the inconsistent observation of mirror-symmetry across prior studies. Pattern analyses of human fMRI data (of either sex) revealed biases compatible with our model. The model provides a unifying explanation of MVPA studies of viewpoint selectivity and suggests observations of mirror-symmetry originate from ineffectively normalized signal imbalances across different face views.

Profiles of visual perceptual learning in feature space

Visual perceptual learning (VPL), experience-induced gains in discriminating visual features, has been studied extensively and intensively for many years, its profile in feature space, however, remains unclear. Here, human subjects were trained to perform either a simple low-level feature (grating orientation) or a complex high-level object (face view) discrimination task over a long-time course. During, immediately after, and one month after training, all results showed that in feature space VPL in grating orientation discrimination was a center-surround profile; VPL in face view discrimination, however, was a monotonic gradient profile. Importantly, these two profiles can be emerged by a deep convolutional neural network with a modified AlexNet consisted of 7 and 12 layers, respectively. Altogether, our study reveals for the first time a feature hierarchy-dependent profile of VPL in feature space, placing a necessary constraint on our understanding of the neural computation of VPL.

Encoding of continuous perceptual choices in human early visual cortex

Introduction

Research on the neural mechanisms of perceptual decision-making has typically focused on simple categorical choices, say between two alternative motion directions. Studies on such discrete alternatives have often suggested that choices are encoded either in a motor-based or in an abstract, categorical format in regions beyond sensory cortex.

Methods

In this study, we used motion stimuli that could vary anywhere between 0° and 360° to assess how the brain encodes choices for features that span the full sensory continuum. We employed a combination of neuroimaging and encoding models based on Gaussian process regression to assess how either stimuli or choices were encoded in brain responses.

Results

We found that single-voxel tuning patterns could be used to reconstruct the trial-by-trial physical direction of motion as well as the participants' continuous choices. Importantly, these continuous choice signals were primarily observed in early visual areas. The tuning properties in this region generalized between choice encoding and stimulus encoding, even for reports that reflected pure guessing.

Discussion

We found only little information related to the decision outcome in regions beyond visual cortex, such as parietal cortex, possibly because our task did not involve differential motor preparation. This could suggest that decisions for continuous stimuli take can place already in sensory brain regions, potentially using similar mechanisms to the sensory recruitment in visual working memory.

Statistical inference on representational geometries

Neuroscience has recently made much progress, expanding the complexity of both neural activity measurements and brain-computational models. However, we lack robust methods for connecting theory and experiment by evaluating our new big models with our new big data. Here, we introduce new inference methods enabling researchers to evaluate and compare models based on the accuracy of their predictions of representational geometries: A good model should accurately predict the distances among the neural population representations (e.g. of a set of stimuli). Our inference methods combine novel 2-factor extensions of crossvalidation (to prevent overfitting to either subjects or conditions from inflating our estimates of model accuracy) and bootstrapping (to enable inferential model comparison with simultaneous generalization to both new subjects and new conditions). We validate the inference methods on data where the ground-truth model is known, by simulating data with deep neural networks and by resampling of calcium-imaging and functional MRI data. Results demonstrate that the methods are valid and conclusions generalize correctly. These data analysis methods are available in an open-source Python toolbox (rsatoolbox.readthedocs.io).

Emergence of brain-like mirror-symmetric viewpoint tuning in convolutional neural networks

Primates can recognize objects despite 3D geometric variations such as in-depth rotations. The computational mechanisms that give rise to such invariances are yet to be fully understood. A curious case of partial invariance occurs in the macaque face-patch AL and in fully connected layers of deep convolutional networks in which neurons respond similarly to mirror-symmetric views (e.g., left and right profiles). Why does this tuning develop? Here, we propose a simple learning-driven explanation for mirror-symmetric viewpoint tuning. We show that mirror-symmetric viewpoint tuning for faces emerges in the fully connected layers of convolutional deep neural networks trained on object recognition tasks, even when the training dataset does not include faces. First, using 3D objects rendered from multiple views as test stimuli, we demonstrate that mirror-symmetric viewpoint tuning in convolutional neural network models is not unique to faces: it emerges for multiple object categories with bilateral symmetry. Second, we show why this invariance emerges in the models. Learning to discriminate among bilaterally symmetric object categories induces reflection-equivariant intermediate representations. AL-like mirror-symmetric tuning is achieved when such equivariant responses are spatially pooled by downstream units with sufficiently large receptive fields. These results explain how mirror-symmetric viewpoint tuning can emerge in neural networks, providing a theory of how they might emerge in the primate brain. Our theory predicts that mirror-symmetric viewpoint tuning can emerge as a consequence of exposure to bilaterally symmetric objects beyond the category of faces, and that it can generalize beyond previously experienced object categories.

A unifying model for discordant and concordant results in human neuroimaging studies of facial viewpoint selectivity

Our ability to recognize faces regardless of viewpoint is a key property of the primate visual system. Traditional theories hold that facial viewpoint is represented by view-selective mechanisms at early visual processing stages and that representations become increasingly tolerant to viewpoint changes in higher-level visual areas. Newer theories, based on single-neuron monkey electrophysiological recordings, suggest an additional intermediate processing stage invariant to mirror-symmetric face views. Consistent with traditional theories, human studies combining neuroimaging and multivariate pattern analysis (MVPA) methods have provided evidence of view-selectivity in early visual cortex. However, contradictory results have been reported in higher-level visual areas concerning the existence in humans of mirror-symmetrically tuned representations. We believe these results reflect low-level stimulus confounds and data analysis choices. To probe for low-level confounds, we analyzed images from two popular face databases. Analyses of mean image luminance and contrast revealed biases across face views described by even polynomials-i.e., mirror-symmetric. To explain major trends across human neuroimaging studies of viewpoint selectivity, we constructed a network model that incorporates three biological constraints: cortical magnification, convergent feedforward projections, and interhemispheric connections. Given the identified low-level biases, we show that a gradual increase of interhemispheric connections across network layers is sufficient to replicate findings of mirror-symmetry in high-level processing stages, as well as view-tuning in early processing stages. Data analysis decisions-pattern dissimilarity measure and data recentering-accounted for the variable observation of mirror-symmetry in late processing stages. The model provides a unifying explanation of MVPA studies of viewpoint selectivity. We also show how common analysis choices can lead to erroneous conclusions.

Two distinct networks containing position-tolerant representations of actions in the human brain

Humans can recognize others' actions in the social environment. This action recognition ability is rarely hindered by the movement of people in the environment. The neural basis of this position tolerance for observed actions is not fully understood. Here, we aimed to identify brain regions capable of generalizing representations of actions across different positions and investigate the representational content of these regions. In a functional magnetic resonance imaging experiment, participants viewed point-light displays of different human actions. Stimuli were presented in either the upper or the lower visual field. Multivariate pattern analysis and a surface-based searchlight approach were employed to identify brain regions that contain position-tolerant action representation: Classifiers were trained with patterns in response to stimuli presented in one position and were tested with stimuli presented in another position. Results showed above-chance classification in the left and right lateral occipitotemporal cortices, right intraparietal sulcus, and right postcentral gyrus. Further analyses exploring the representational content of these regions showed that responses in the lateral occipitotemporal regions were more related to subjective judgments, while those in the parietal regions were more related to objective measures. These results provide evidence for two networks that contain abstract representations of human actions with distinct representational content.

Improving the validity of neuroimaging decoding tests of invariant and configural neural representation

Many research questions in sensory neuroscience involve determining whether the neural representation of a stimulus property is invariant or specific to a particular stimulus context (e.g., Is object representation invariant to translation? Is the representation of a face feature specific to the context of other face features?). Between these two extremes, representations may also be context-tolerant or context-sensitive. Most neuroimaging studies have used operational tests in which a target property is inferred from a significant test against the null hypothesis of the opposite property. For example, the popular cross-classification test concludes that representations are invariant or tolerant when the null hypothesis of specificity is rejected. A recently developed neurocomputational theory suggests two insights regarding such tests. First, tests against the null of context-specificity, and for the alternative of context-invariance, are prone to false positives due to the way in which the underlying neural representations are transformed into indirect measurements in neuroimaging studies. Second, jointly performing tests against the nulls of invariance and specificity allows one to reach more precise and valid conclusions about the underlying representations, particularly when the null of invariance is tested using the fine-grained information from classifier decision variables rather than only accuracies (i.e., using the decoding separability test). Here, we provide empirical and computational evidence supporting both of these theoretical insights. In our empirical study, we use encoding of orientation and spatial position in primary visual cortex as a case study, as previous research has established that these properties are encoded in a context-sensitive way. Using fMRI decoding, we show that the cross-classification test produces false-positive conclusions of invariance, but that more valid conclusions can be reached by jointly performing tests against the null of invariance. The results of two simulations further support both of these conclusions. We conclude that more valid inferences about invariance or specificity of neural representations can be reached by jointly testing against both hypotheses, and using neurocomputational theory to guide the interpretation of results.

A modeling framework for determining modulation of neural-level tuning from non-invasive human fMRI data

Many neuroscience theories assume that tuning modulation of individual neurons underlies changes in human cognition. However, non-invasive fMRI lacks sufficient resolution to visualize this modulation. To address this limitation, we developed an analysis framework called Inferring Neural Tuning Modulation (INTM) for "peering inside" voxels. Precise specification of neural tuning from the BOLD signal is not possible. Instead, INTM compares theoretical alternatives for the form of neural tuning modulation that might underlie changes in BOLD across experimental conditions. The most likely form is identified via formal model comparison, with assumed parametric Normal tuning functions, followed by a non-parametric check of conclusions. We validated the framework by successfully identifying a well-established form of modulation: visual contrast-induced multiplicative gain for orientation tuned neurons. INTM can be applied to any experimental paradigm testing several points along a continuous feature dimension (e.g., direction of motion, isoluminant hue) across two conditions (e.g., with/without attention, before/after learning).

The emergence of view-symmetric neural responses to familiar and unfamiliar faces

Successful recognition of familiar faces is thought to depend on the ability to integrate view-dependent representations of a face into a view-invariant representation. It has been proposed that a key intermediate step in achieving view invariance is the representation of symmetrical views. However, key unresolved questions remain, such as whether these representations are specific for naturally occurring changes in viewpoint and whether view-symmetric representations exist for familiar faces. To address these issues, we compared behavioural and neural responses to natural (canonical) and unnatural (noncanonical) rotations of the face. Similarity judgements revealed that symmetrical viewpoints were perceived to be more similar than non-symmetrical viewpoints for both canonical and non-canonical rotations. Next, we measured patterns of neural response from early to higher level regions of visual cortex. Early visual areas showed a view-dependent representation for natural or canonical rotations of the face, such that the similarity between patterns of response were related to the difference in rotation. View symmetric patterns of neural response to canonically rotated faces emerged in higher visual areas, particularly in face-selective regions. The emergence of a view-symmetric representation from a view-dependent representation for canonical rotations of the face was also evident for familiar faces, suggesting that view-symmetry is an important intermediate step in generating view-invariant representations. Finally, we measured neural responses to unnatural or non-canonical rotations of the face. View-symmetric patterns of response were also found in face-selective regions. However, in contrast to natural or canonical rotations of the face, these view-symmetric responses did not arise from an initial view-dependent representation in early visual areas. This suggests differences in the way that view-symmetrical representations emerge with canonical or non-canonical rotations. The similarity in the neural response to canonical views of familiar and unfamiliar faces in the core face network suggests that the neural correlates of familiarity emerge at later stages of processing.

The effects of stimulus inversion on the neural representations of Chinese character and face recognition

This study investigates whether stimulus inversion influences neural responses of Chinese character recognition similarly to its effect on face recognition in category-selective and object-related brain areas using functional magnetic resonance imaging. Participants performed a one-back matching task for simple (one radical) and compound (two radicals) Chinese characters and faces with upright and inverted orientations. Inverted stimuli produced slower response times with stronger activity within the fusiform gyrus (FG) than upright stimuli for faces and Chinese characters. While common inversion-related activation was identified in the left FG among stimulus types, we observed a significant inter-regional correlation between the left FG and the intraparietal sulcus for face inversion. Importantly, analyses of region-of-interest (ROI) multivariate pattern classification showed that classifiers trained on face inversion can decode the representations of character inversion in the character-selective ROI. However, this was not true for face inversion in face-selective ROIs when the classifiers were trained on characters. Similar activity patterns for character and face inversion were observed in the object-related ROIs. We also showed higher decoding accuracy for upright stimuli in the face-selective ROI than in the character-selective ROI but this was not true for inverted ones or when patterns were examined in the object-related ROIs. Together, our results support shared and distinct configural representations for character and face recognition in category-selective and object-related brain areas.

A Distributed Model of Face and Body Integration

Separated face- and body-responsive brain networks have been identified that show strong responses when observers view faces and bodies. It has been proposed that face and body processing may be initially separated in the lateral occipitotemporal cortex and then combined into a whole person representation in the anterior temporal cortex, or elsewhere in the brain. However, in contrast to this proposal, our recent study identified a common coding of face and body orientation (ie, facing direction) in the lateral occipitotemporal cortex, demonstrating an integration of face and body information at an early stage of face and body processing. These results, in combination with findings that show integration of face and body identity in the lateral occipitotemporal, parahippocampal and superior parietal cortex, and face and body emotional expression in the posterior superior temporal sulcus and medial prefrontal cortex, suggest that face and body integration may be more distributed than previously considered. I propose a new model of face and body integration, where areas at the intersection of face- and body-responsive regions play a role in integrating specific properties of faces and bodies, and distributed regions across the brain contribute to high-level, abstract integration of shared face and body properties.

The neural coding of face and body orientation in occipitotemporal cortex

Face and body orientation convey important information for us to understand other people's actions, intentions and social interactions. It has been shown that several occipitotemporal areas respond differently to faces or bodies of different orientations. However, whether face and body orientation are processed by partially overlapping or completely separate brain networks remains unclear, as the neural coding of face and body orientation is often investigated separately. Here, we recorded participants' brain activity using fMRI while they viewed faces and bodies shown from three different orientations, while attending to either orientation or identity information. Using multivoxel pattern analysis we investigated which brain regions process face and body orientation respectively, and which regions encode both face and body orientation in a stimulus-independent manner. We found that patterns of neural responses evoked by different stimulus orientations in the occipital face area, extrastriate body area, lateral occipital complex and right early visual cortex could generalise across faces and bodies, suggesting a stimulus-independent encoding of person orientation in occipitotemporal cortex. This finding was consistent across functionally defined regions of interest and a whole-brain searchlight approach. The fusiform face area responded to face but not body orientation, suggesting that orientation responses in this area are face-specific. Moreover, neural responses to orientation were remarkably consistent regardless of whether participants attended to the orientation of faces and bodies or not. Together, these results demonstrate that face and body orientation are processed in a partially overlapping brain network, with a stimulus-independent neural code for face and body orientation in occipitotemporal cortex.

The roles of gaze and head orientation in face categorization during rapid serial visual presentation

Little is known about how perceived gaze direction and head orientation may influence human categorization of visual stimuli as faces. To address this question, a sequence of unsegmented natural images, each containing a random face or a non-face object, was presented in rapid succession (stimulus duration: 91.7 ms per image) during which human observers were instructed to respond immediately to every face presentation. Faces differed in gaze and head orientation in 7 combinations - full-front views with perceived gaze (1) directed to the observer, (2) averted to the left, or (3) averted to the right, left ¾ side views with (4) direct gaze or (5) averted gaze, and right ¾ side views with (6) direct gaze or (7) averted gaze - were presented randomly throughout the sequence. We found highly accurate and rapid behavioural responses to all kinds of faces. Crucially, both perceived gaze direction and head orientation had comparable, non-interactive effects on response times, where direct gaze was responded faster than averted gaze by 48 ms and full-front view faster than ¾ side view also by 48 ms on average. Presentations of full-front faces with direct gaze led to an additive speed advantage of 96 ms to ¾ faces with averted gaze. The results reveal that the effects of perceived gaze direction and head orientation on the speed of face categorization probably depend on the degree of social relevance of the face to the viewer.

Separated and overlapping neural coding of face and body identity

Recognising a person's identity often relies on face and body information, and is tolerant to changes in low‐level visual input (e.g., viewpoint changes). Previous studies have suggested that face identity is disentangled from low‐level visual input in the anterior face‐responsive regions. It remains unclear which regions disentangle body identity from variations in viewpoint, and whether face and body identity are encoded separately or combined into a coherent person identity representation. We trained participants to recognise three identities, and then recorded their brain activity using fMRI while they viewed face and body images of these three identities from different viewpoints. Participants' task was to respond to either the stimulus identity or viewpoint. We found consistent decoding of body identity across viewpoint in the fusiform body area, right anterior temporal cortex, middle frontal gyrus and right insula. This finding demonstrates a similar function of fusiform and anterior temporal cortex for bodies as has previously been shown for faces, suggesting these regions may play a general role in extracting high‐level identity information. Moreover, we could decode identity across fMRI activity evoked by faces and bodies in the early visual cortex, right inferior occipital cortex, right parahippocampal cortex and right superior parietal cortex, revealing a distributed network that encodes person identity abstractly. Lastly, identity decoding was consistently better when participants attended to identity, indicating that attention to identity enhances its neural representation. These results offer new insights into how the brain develops an abstract neural coding of person identity, shared by faces and bodies.

Does automatic human face categorization depend on head orientation?

Whether human categorization of visual stimuli as faces is optimal for full-front views, best revealing diagnostic features but lacking depth cues, remains largely unknown. To address this question, we presented 16 human observers with unsegmented natural images of different living and non-living objects at a fast rate (f = 12 Hz), with natural face images appearing at f/9 = 1.33 Hz. Faces posing all full-front or at ¾ side view angles appeared in separate sequences. Robust frequency-tagged 1.33 Hz (and harmonic) occipito-temporal electroencephalographic (EEG) responses reflecting face-selective neural activity did not differ in overall amplitude between full-front and ¾ side views. Despite this, alternating between full-front and ¾ side views within a sequence led to significant responses at specific harmonics of .67 Hz (f/18), objectively isolating view-dependent face-selective responses over occipito-temporal regions. Critically, a time-domain analysis showed that these view-dependent face-selective responses reflected only an earlier response to full-front than ¾ side views by 8-13 ms. Overall, these findings indicate that the face-selective neural representation is as robust for ¾ side faces as for full-front faces in the human brain, but full-front views provide a slightly earlier processing-time advantage as compared to rotated face views.

FFA and OFA Encode Distinct Types of Face Identity Information

Faces of different people elicit distinct fMRI patterns in several face-selective regions of the human brain. Here we used representational similarity analysis to investigate what type of identity-distinguishing information is encoded in three face-selective regions: fusiform face area (FFA), occipital face area (OFA), and posterior superior temporal sulcus (pSTS). In a sample of 30 human participants (22 females, 8 males), we used fMRI to measure brain activity patterns elicited by naturalistic videos of famous face identities, and compared their representational distances in each region with models of the differences between identities. We built diverse candidate models, ranging from low-level image-computable properties (pixel-wise, GIST, and Gabor-Jet dissimilarities), through higher-level image-computable descriptions (OpenFace deep neural network, trained to cluster faces by identity), to complex human-rated properties (perceived similarity, social traits, and gender). We found marked differences in the information represented by the FFA and OFA. Dissimilarities between face identities in FFA were accounted for by differences in perceived similarity, Social Traits, Gender, and by the OpenFace network. In contrast, representational distances in OFA were mainly driven by differences in low-level image-based properties (pixel-wise and Gabor-Jet dissimilarities). Our results suggest that, although FFA and OFA can both discriminate between identities, the FFA representation is further removed from the image, encoding higher-level perceptual and social face information.SIGNIFICANCE STATEMENT Recent studies using fMRI have shown that several face-responsive brain regions can distinguish between different face identities. It is however unclear whether these different face-responsive regions distinguish between identities in similar or different ways. We used representational similarity analysis to investigate the computations within three brain regions in response to naturalistically varying videos of face identities. Our results revealed that two regions, the fusiform face area and the occipital face area, encode distinct identity information about faces. Although identity can be decoded from both regions, identity representations in fusiform face area primarily contained information about social traits, gender, and high-level visual features, whereas occipital face area primarily represented lower-level image features.

Effects of the Openness to Experience Polygenic Score on Cortical Thickness and Functional Connectivity

Openness to experience (OTE) has relatively stable and heritable characteristics. Previous studies have used candidate gene approaches to explore the genetic mechanisms of OTE, but genome-wide polygenic scores have a greater genetic effect than other genetic analysis methods, and previous studies have never examined the potential effect of OTE on this cumulative effect at the level of the brain mechanism. In the present study, we aim to explore the associations between polygenic scores (PGSs) of OTE and brain structure and functions. First, the results of PGSs of OTE at seven different thresholds were calculated in a large Chinese sample (N = 586). Then, we determined the associations between PGSs of OTE and cortical thickness and functional connectivity. The results showed that PGSs of OTE was negatively correlated with the thickness of the fusiform gyrus, and PGSs of OTE were negatively associated with the functional connectivity between the left intraparietal sulcus (IPS) and the right posterior occipital lobe. These findings may suggest that the brain structure of fusiform gyrus and brain functions of IPS and posterior occipital lobe are partly regulated by OTE-related genetic factors.

What do across-subject analyses really tell us about neural coding?

A key challenge in human neuroscience is to gain information about patterns of neural activity using indirect measures. Multivariate pattern analysis methods testing for generalization of information across subjects have been used to support inferences regarding neural coding. One critical assumption of an important class of such methods is that anatomical normalization is suited to align spatially-structured neural patterns across individual brains. We asked whether anatomical normalization is suited for this purpose. If not, what sources of information are such across-subject cross-validated analyses likely to reveal? To investigate these questions, we implemented two-layered feedforward randomly-connected networks. A key feature of these simulations was a gain-field with a spatial structure shared across networks. To investigate whether total-signal imbalances across conditions-e.g. differences in overall activity-affect the observed pattern of results, we manipulated the energy-profile of images conforming to a pre-specified correlation structure. To investigate whether the level of granularity of the data also influences results, we manipulated the density of connections between network layers. Simulations showed that anatomical normalization is unsuited to align neural representations. Pattern similarity-relationships were explained by the observed total-signal imbalances across conditions. Further, we observed that deceptively complex representational structures emerge from arbitrary analysis choices, such as whether the data are mean-subtracted during preprocessing. These simulations also led to testable predictions regarding the distribution of low-level features in images used in recent fMRI studies that relied on leave-one-subject-out pattern analyses. Image analyses broadly confirmed these predictions. Finally, hyperalignment emerged as a principled alternative to test across-subject generalization of spatially-structured information. We illustrate cases in which hyperalignment proved successful, as well as cases in which it only partially recovered the latent correlation structure in the pattern of responses. Our results highlight the need for robust, high-resolution measurements from individual subjects. We also offer a way forward for across-subject analyses. We suggest ways to inform hyperalignment results with estimates of the strength of the signal associated with each condition. Such information can usefully constrain ensuing inferences regarding latent representational structures as well as population tuning dimensions.

Face Representations via Tensorfaces of Various Complexities

Neurons selective for faces exist in humans and monkeys. However, characteristics of face cell receptive fields are poorly understood. In this theoretical study, we explore the effects of complexity, defined as algorithmic information (Kolmogorov complexity) and logical depth, on possible ways that face cells may be organized. We use tensor decompositions to decompose faces into a set of components, called tensorfaces, and their associated weights, which can be interpreted as model face cells and their firing rates. These tensorfaces form a high-dimensional representation space in which each tensorface forms an axis of the space. A distinctive feature of the decomposition algorithm is the ability to specify tensorface complexity. We found that low-complexity tensorfaces have blob-like appearances crudely approximating faces, while high-complexity tensorfaces appear clearly face-like. Low-complexity tensorfaces require a larger population to reach a criterion face reconstruction error than medium- or high-complexity tensorfaces, and thus are inefficient by that criterion. Low-complexity tensorfaces, however, generalize better when representing statistically novel faces, which are faces falling beyond the distribution of face description parameters found in the tensorface training set. The degree to which face representations are parts based or global forms a continuum as a function of tensorface complexity, with low and medium tensorfaces being more parts based. Given the computational load imposed in creating high-complexity face cells (in the form of algorithmic information and logical depth) and in the absence of a compelling advantage to using high-complexity cells, we suggest face representations consist of a mixture of low- and medium-complexity face cells.

Measures of Neural Similarity

One fundamental question is what makes two brain states similar. For example, what makes the activity in visual cortex elicited from viewing a robin similar to a sparrow? One common assumption in fMRI analysis is that neural similarity is described by Pearson correlation. However, there are a host of other possibilities, including Minkowski and Mahalanobis measures, with each differing in its mathematical, theoretical, and neural computational assumptions. Moreover, the operable measures may vary across brain regions and tasks. Here, we evaluated which of several competing similarity measures best captured neural similarity. Our technique uses a decoding approach to assess the information present in a brain region, and the similarity measures that best correspond to the classifier’s confusion matrix are preferred. Across two published fMRI datasets, we found the preferred neural similarity measures were common across brain regions but differed across tasks. Moreover, Pearson correlation was consistently surpassed by alternatives.

Representational structure or task structure? Bias in neural representational similarity analysis and a Bayesian method for reducing bias

The activity of neural populations in the brains of humans and animals can exhibit vastly different spatial patterns when faced with different tasks or environmental stimuli. The degrees of similarity between these neural activity patterns in response to different events are used to characterize the representational structure of cognitive states in a neural population. The dominant methods of investigating this similarity structure first estimate neural activity patterns from noisy neural imaging data using linear regression, and then examine the similarity between the estimated patterns. Here, we show that this approach introduces spurious bias structure in the resulting similarity matrix, in particular when applied to fMRI data. This problem is especially severe when the signal-to-noise ratio is low and in cases where experimental conditions cannot be fully randomized in a task. We propose Bayesian Representational Similarity Analysis (BRSA), an alternative method for computing representational similarity, in which we treat the covariance structure of neural activity patterns as a hyper-parameter in a generative model of the neural data. By marginalizing over the unknown activity patterns, we can directly estimate this covariance structure from imaging data. This method offers significant reductions in bias and allows estimation of neural representational similarity with previously unattained levels of precision at low signal-to-noise ratio, without losing the possibility of deriving an interpretable distance measure from the estimated similarity. The method is closely related to Pattern Component Model (PCM), but instead of modeling the estimated neural patterns as in PCM, BRSA models the imaging data directly and is suited for analyzing data in which the order of task conditions is not fully counterbalanced. The probabilistic framework allows for jointly analyzing data from a group of participants. The method can also simultaneously estimate a signal-to-noise ratio map that shows where the learned representational structure is supported more strongly. Both this map and the learned covariance matrix can be used as a structured prior for maximum a posteriori estimation of neural activity patterns, which can be further used for fMRI decoding. Our method therefore paves the way towards a more unified and principled analysis of neural representations underlying fMRI signals. We make our tool freely available in Brain Imaging Analysis Kit (BrainIAK).

Symmetrical Viewpoint Representations in Face-Selective Regions Convey an Advantage in the Perception and Recognition of Faces

Learning new identities is crucial for effective social interaction. A critical aspect of this process is the integration of different images from the same face into a view-invariant representation that can be used for recognition. The representation of symmetrical viewpoints has been proposed to be a key computational step in achieving view-invariance. The aim of this study was to determine whether the representation of symmetrical viewpoints in face-selective regions is directly linked to the perception and recognition of face identity. In Experiment 1, we measured fMRI responses while male and female human participants viewed images of real faces from different viewpoints (-90, -45, 0, 45, and 90° from full-face view). Within the face regions, patterns of neural response to symmetrical views (-45 and 45° or -90 and 90°) were more similar than responses to nonsymmetrical views in the fusiform face area and superior temporal sulcus, but not in the occipital face area. In Experiment 2, participants made perceptual similarity judgements to pairs of face images. Images with symmetrical viewpoints were reported as being more similar than nonsymmetric views. In Experiment 3, we asked whether symmetrical views also convey an advantage when learning new faces. We found that recognition was best when participants were tested with novel face images that were symmetrical to the learning viewpoint. Critically, the pattern of perceptual similarity and recognition across different viewpoints predicted the pattern of neural response in face-selective regions. Together, our results provide support for the functional value of symmetry as an intermediate step in generating view-invariant representations.SIGNIFICANCE STATEMENT The recognition of identity from faces is crucial for successful social interactions. A critical step in this process is the integration of different views into a unified, view-invariant representation. The representation of symmetrical views (e.g., left profile and right profile) has been proposed as an important intermediate step in computing view-invariant representations. We found view symmetric representations were specific to some face-selective regions, but not others. We also show that these neural representations influence the perception of faces. Symmetric views were perceived to be more similar and were recognized more accurately than nonsymmetric views. Moreover, the perception and recognition of faces at different viewpoints predicted patterns of response in those face regions with view symmetric representations.

Neural coding: A single neuron's perspective

What any sensory neuron knows about the world is one of the cardinal questions in Neuroscience. Information from the sensory periphery travels across synaptically coupled neurons as each neuron encodes information by varying the rate and timing of its action potentials (spikes). Spatiotemporally correlated changes in this spiking regimen across neuronal populations are the neural basis of sensory representations. In the somatosensory cortex, however, spiking of individual (or pairs of) cortical neurons is only minimally informative about the world. Recent studies showed that one solution neurons implement to counteract this information loss is adapting their rate of information transfer to the ongoing synaptic activity by changing the membrane potential at which spike is generated. Here we first introduce the principles of information flow from the sensory periphery to the primary sensory cortex in a model sensory (whisker) system, and subsequently discuss how the adaptive spike threshold gates the intracellular information transfer from the somatic post-synaptic potential to action potentials, controlling the information content of communication across somatosensory cortical neurons.

Orientation Encoding and Viewpoint Invariance in Face Recognition: Inferring Neural Properties from Large-Scale Signals

Viewpoint-invariant face recognition is thought to be subserved by a distributed network of occipitotemporal face-selective areas that, except for the human anterior temporal lobe, have been shown to also contain face-orientation information. This review begins by highlighting the importance of bilateral symmetry for viewpoint-invariant recognition and face-orientation perception. Then, monkey electrophysiological evidence is surveyed describing key tuning properties of face-selective neurons-including neurons bimodally tuned to mirror-symmetric face-views-followed by studies combining functional magnetic resonance imaging (fMRI) and multivariate pattern analyses to probe the representation of face-orientation and identity information in humans. Altogether, neuroimaging studies suggest that face-identity is gradually disentangled from face-orientation information along the ventral visual processing stream. The evidence seems to diverge, however, regarding the prevalent form of tuning of neural populations in human face-selective areas. In this context, caveats possibly leading to erroneous inferences regarding mirror-symmetric coding are exposed, including the need to distinguish angular from Euclidean distances when interpreting multivariate pattern analyses. On this basis, this review argues that evidence from the fusiform face area is best explained by a view-sensitive code reflecting head angular disparity, consistent with a role of this area in face-orientation perception. Finally, the importance is stressed of explicit models relating neural properties to large-scale signals.

The neural basis of visual symmetry and its role in mid- and high-level visual processing

Symmetry is an important and prominent feature of the visual world. It has been studied as a basis for image segmentation and perceptual organization, but it also plays a role in higher level processes, such as face and object perception. Over the past decade, there has been progress in the study of the neural mechanisms of symmetry perception in humans and other animals. There is extended activity in the ventral stream, including the lateral occipital complex (LOC) and VO1; this activity starts in V3 and it occurs independently of the task (automatic response). Additionally, when the task requires processing of symmetry, the activation may emerge for objects that are symmetrical, even though they do not project a symmetrical image. There is also some evidence of hemispheric lateralization, especially for the LOC. We review the studies on the cortical basis of visual symmetry processing and its links to encoding of other aspects of the visual world, such as faces and objects.

Patterns of neural response in face regions are predicted by low-level image properties

Models of face processing suggest that the neural response in different face regions is selective for higher-level attributes of the face, such as identity and expression. However, it remains unclear to what extent the response in these regions can also be explained by more basic organizing principles. Here, we used functional magnetic resonance imaging multivariate pattern analysis (fMRI-MVPA) to ask whether spatial patterns of response in the core face regions (occipital face area - OFA, fusiform face area - FFA, superior temporal sulcus - STS) can be predicted across different participants by lower level properties of the stimulus. First, we compared the neural response to face identity and viewpoint, by showing images of different identities from different viewpoints. The patterns of neural response in the core face regions were predicted by the viewpoint, but not the identity of the face. Next, we compared the neural response to viewpoint and expression, by showing images with different expressions from different viewpoints. Again, viewpoint, but not expression, predicted patterns of response in face regions. Finally, we show that the effect of viewpoint in both experiments could be explained by changes in low-level image properties. Our results suggest that a key determinant of the neural representation in these core face regions involves lower-level image properties rather than an explicit representation of higher-level attributes in the face. The advantage of a relatively image-based representation is that it can be used flexibly in the perception of faces.

Manifold decoding for neural representations of face viewpoint and gaze direction using magnetoencephalographic data

The main challenge in decoding neural representations lies in linking neural activity to representational content or abstract concepts. The transformation from a neural-based to a low-dimensional representation may hold the key to encoding perceptual processes in the human brain. In this study, we developed a novel model by which to represent two changeable features of faces: face viewpoint and gaze direction. These features are embedded in spatiotemporal brain activity derived from magnetoencephalographic data. Our decoding results demonstrate that face viewpoint and gaze direction can be represented by manifold structures constructed from brain responses in the bilateral occipital face area and right superior temporal sulcus, respectively. Our results also show that the superposition of brain activity in the manifold space reveals the viewpoints of faces as well as directions of gazes as perceived by the subject. The proposed manifold representation model provides a novel opportunity to gain further insight into the processing of information in the human brain.

Embedding Anatomical or Functional Knowledge in Whole-Brain Multiple Kernel Learning Models

Pattern recognition models have been increasingly applied to neuroimaging data over the last two decades. These applications have ranged from cognitive neuroscience to clinical problems. A common limitation of these approaches is that they do not incorporate previous knowledge about the brain structure and function into the models. Previous knowledge can be embedded into pattern recognition models by imposing a grouping structure based on anatomically or functionally defined brain regions. In this work, we present a novel approach that uses group sparsity to model the whole brain multivariate pattern as a combination of regional patterns. More specifically, we use a sparse version of Multiple Kernel Learning (MKL) to simultaneously learn the contribution of each brain region, previously defined by an atlas, to the decision function. Our application of MKL provides two beneficial features: (1) it can lead to improved overall generalisation performance when the grouping structure imposed by the atlas is consistent with the data; (2) it can identify a subset of relevant brain regions for the predictive model. In order to investigate the effect of the grouping in the proposed MKL approach we compared the results of three different atlases using three different datasets. The method has been implemented in the new version of the open-source Pattern Recognition for Neuroimaging Toolbox (PRoNTo).

Adjudicating between face-coding models with individual-face fMRI responses

The perceptual representation of individual faces is often explained with reference to a norm-based face space. In such spaces, individuals are encoded as vectors where identity is primarily conveyed by direction and distinctiveness by eccentricity. Here we measured human fMRI responses and psychophysical similarity judgments of individual face exemplars, which were generated as realistic 3D animations using a computer-graphics model. We developed and evaluated multiple neurobiologically plausible computational models, each of which predicts a representational distance matrix and a regional-mean activation profile for 24 face stimuli. In the fusiform face area, a face-space coding model with sigmoidal ramp tuning provided a better account of the data than one based on exemplar tuning. However, an image-processing model with weighted banks of Gabor filters performed similarly. Accounting for the data required the inclusion of a measurement-level population averaging mechanism that approximates how fMRI voxels locally average distinct neuronal tunings. Our study demonstrates the importance of comparing multiple models and of modeling the measurement process in computational neuroimaging.

Representational models: A common framework for understanding encoding, pattern-component, and representational-similarity analysis

Representational models specify how activity patterns in populations of neurons (or, more generally, in multivariate brain-activity measurements) relate to sensory stimuli, motor responses, or cognitive processes. In an experimental context, representational models can be defined as hypotheses about the distribution of activity profiles across experimental conditions. Currently, three different methods are being used to test such hypotheses: encoding analysis, pattern component modeling (PCM), and representational similarity analysis (RSA). Here we develop a common mathematical framework for understanding the relationship of these three methods, which share one core commonality: all three evaluate the second moment of the distribution of activity profiles, which determines the representational geometry, and thus how well any feature can be decoded from population activity. Using simulated data for three different experimental designs, we compare the power of the methods to adjudicate between competing representational models. PCM implements a likelihood-ratio test and therefore provides the most powerful test if its assumptions hold. However, the other two approaches-when conducted appropriately-can perform similarly. In encoding analysis, the linear model needs to be appropriately regularized, which effectively imposes a prior on the activity profiles. With such a prior, an encoding model specifies a well-defined distribution of activity profiles. In RSA, the unequal variances and statistical dependencies of the dissimilarity estimates need to be taken into account to reach near-optimal power in inference. The three methods render different aspects of the information explicit (e.g. single-response tuning in encoding analysis and population-response representational dissimilarity in RSA) and have specific advantages in terms of computational demands, ease of use, and extensibility. The three methods are properly construed as complementary components of a single data-analytical toolkit for understanding neural representations on the basis of multivariate brain-activity data.

Representational Dynamics of Facial Viewpoint Encoding

Faces provide a wealth of information, including the identity of the seen person and social cues, such as the direction of gaze. Crucially, different aspects of face processing require distinct forms of information encoding. Another person's attentional focus can be derived based on a view-dependent code. In contrast, identification benefits from invariance across all viewpoints. Different cortical areas have been suggested to subserve these distinct functions. However, little is known about the temporal aspects of differential viewpoint encoding in the human brain. Here, we combine EEG with multivariate data analyses to resolve the dynamics of face processing with high temporal resolution. This revealed a distinct sequence of viewpoint encoding. Head orientations were encoded first, starting after around 60 msec of processing. Shortly afterward, peaking around 115 msec after stimulus onset, a different encoding scheme emerged. At this latency, mirror-symmetric viewing angles elicited highly similar cortical responses. Finally, about 280 msec after visual onset, EEG response patterns demonstrated a considerable degree of viewpoint invariance across all viewpoints tested, with the noteworthy exception of the front-facing view. Taken together, our results indicate that the processing of facial viewpoints follows a temporal sequence of encoding schemes, potentially mirroring different levels of computational complexity.

A Primer on Pattern-Based Approaches to fMRI: Principles, Pitfalls, and Perspectives

Human fMRI signals exhibit a spatial patterning that contains detailed information about a person's mental states. Using classifiers it is possible to access this information and study brain processes at the level of individual mental representations. The precise link between fMRI signals and neural population signals still needs to be unraveled. Also, the interpretation of classification studies needs to be handled with care. Nonetheless, pattern-based analyses make it possible to investigate human representational spaces in unprecedented ways, especially when combined with computational modeling.