Data leakage in deep learning studies of translational EEG.

A growing number of studies apply deep neural networks (DNNs) to recordings of human electroencephalography (EEG) to identify a range of disorders. In many studies, EEG recordings are split into segments, and each segment is randomly assigned to the training or test set. As a consequence, data from individual subjects appears in both the training and the test set. Could high test-set accuracy reflect data leakage from subject-specific patterns in the data, rather than patterns that identify a disease? We address this question by testing the performance of DNN classifiers using segment-based holdout (in which segments from one subject can appear in both the training and test set), and comparing this to their performance using subject-based holdout (where all segments from one subject appear exclusively in either the training set or the test set). In two datasets (one classifying Alzheimer's disease, and the other classifying epileptic seizures), we find that performance on previously-unseen subjects is strongly overestimated when models are trained using segment-based holdout. Finally, we survey the literature and find that the majority of translational DNN-EEG studies use segment-based holdout. Most published DNN-EEG studies may dramatically overestimate their classification performance on new subjects.

Visual word processing engages a hierarchical, distributed, and bilateral cortical network.

Although the Visual Word Form Area (VWFA) in left temporal cortex is considered the pre-eminent region in visual word processing, other regions are also implicated. We examined the entire text-selective circuit, using functional MRI. Ten regions of interest (ROIs) per hemisphere were defined, which, based on clustering, grouped into early vision, high-level vision, and language clusters. We analyzed the responses of the ROIs and clusters to words, inverted words, and consonant strings using univariate, multivariate, and functional connectivity measures. Bilateral modulation by stimulus condition was evident, with a stronger effect in left hemisphere regions. Last, using graph theory, we observed that the VWFA was equivalently connected with early visual and language clusters in both hemispheres, reflecting its role as a mediator in the circuit. Although the individual ROIs and clusters bilaterally were flexibly altered by the nature of the input, stability held at the level of global circuit connectivity, reflecting the complex hierarchical distributed system serving visual text perception.

A connectivity-constrained computational account of topographic organization in primate high-level visual cortex.

Inferotemporal (IT) cortex in humans and other primates is topographically organized, containing multiple hierarchically organized areas selective for particular domains, such as faces and scenes. This organization is commonly viewed in terms of evolved domain-specific visual mechanisms. Here, we develop an alternative, domain-general and developmental account of IT cortical organization. The account is instantiated in interactive topographic networks (ITNs), a class of computational models in which a hierarchy of model IT areas, subject to biologically plausible connectivity-based constraints, learns high-level visual representations optimized for multiple domains. We find that minimizing a wiring cost on spatially organized feedforward and lateral connections, alongside realistic constraints on the sign of neuronal connectivity within model IT, results in a hierarchical, topographic organization. This organization replicates a number of key properties of primate IT cortex, including the presence of domain-selective spatial clusters preferentially involved in the representation of faces, objects, and scenes; columnar responses across separate excitatory and inhibitory units; and generic spatial organization whereby the response correlation of pairs of units falls off with their distance. We thus argue that topographic domain selectivity is an emergent property of a visual system optimized to maximize behavioral performance under generic connectivity-based constraints.

Deep learning of shared perceptual representations for familiar and unfamiliar faces: Reply to commentaries.

We recently argued that human unfamiliar face identity perception reflects substantial perceptual expertise, and that the advantage for familiar over unfamiliar face identity matching reflects a learned mapping between generic high-level perceptual features and a unique identity representation of each individual (Blauch, Behrmann and Plaut, 2020). Here we respond to two commentaries by Young and Burton (2020) and Yovel and Abudarham (2020), clarifying and elaborating our stance on various theoretical issues, and discussing topics for future research in human face recognition and the learning of perceptual representations.

Computational insights into human perceptual expertise for familiar and unfamiliar face recognition.

Humans are generally thought to be experts at face recognition, and yet identity perception for unfamiliar faces is surprisingly poor compared to that for familiar faces. Prior theoretical work has argued that unfamiliar face identity perception suffers because the majority of identity-invariant visual variability is idiosyncratic to each identity, and thus, each face identity must be learned essentially from scratch. Using a high-performing deep convolutional neural network, we evaluate this claim by examining the effects of visual experience in untrained, object-expert and face-expert networks. We found that only face training led to substantial generalization in an identity verification task of novel unfamiliar identities. Moreover, generalization increased with the number of previously learned identities, highlighting the generality of identity-invariant information in face images. To better understand how familiarity builds upon generic face representations, we simulated familiarization with face identities by fine-tuning the network on images of the previously unfamiliar identities. Familiarization produced a sharp boost in verification, but only approached ceiling performance in the networks that were highly trained on faces. Moreover, in these face-expert networks, the sharp familiarity benefit was seen only at the identity-based output probability layer, and did not depend on changes to perceptual representations; rather, familiarity effects required learning only at the level of identity readout from a fixed expert representation. Our results thus reconcile the existence of a large familiar face advantage with claims that both familiar and unfamiliar face identity processing depend on shared expert perceptual representations.

Uncharacteristic Task-Evoked Pupillary Responses Implicate Atypical Locus Ceruleus Activity in Autism.

Autism spectrum disorder (ASD) is characterized partly by atypical attentional engagement, reflected in exaggerated and variable responses to sensory stimuli. Attentional engagement is known to be regulated by the locus ceruleus (LC). Moderate baseline LC activity globally dampens neural responsivity and is associated with adaptive deployment and narrowing of attention to task-relevant stimuli. In contrast, increased baseline LC activity enhances neural responsivity across cortex and widening of attention to environmental stimuli regardless of their task relevance. Given attentional atypicalities in ASD, this study is the first to evaluate whether, under different attentional task demands, individuals with ASD exhibit a different profile of LC activity compared with typically developing controls. Males and females with ASD and age- and gender-matched controls participated in a one-back letter detection test while task-evoked pupillary responses, an established correlate for LC activity, were recorded. Participants completed this task in two conditions, either in the absence or presence of distractor auditory tones. Compared with controls, individuals with ASD evinced atypical pupillary responses in the presence versus absence of distractors. Notably, this atypical pupillary profile was evident despite the fact that both groups exhibited equivalent task performance. Moreover, between-group differences in pupillary responses were observed specifically in response to task-relevant events, providing confirmation that the group differences most likely were specifically associated with distinctions in LC activity. These findings suggest that individuals with ASD show atypical modulation of LC activity with changes in attentional demands, offering a possible mechanistic and neurobiological account for attentional atypicalities in ASD.SIGNIFICANCE STATEMENT Individuals with autism spectrum disorder (ASD) exhibit atypical attentional behaviors, including altered sensory responses and atypical fixedness, but the neural mechanism underlying these behaviors remains elusive. One candidate mechanism is atypical locus ceruleus (LC) activity, as the LC plays a critical role in attentional modulation. Specifically, LC activity is involved in regulating the trade-off between environmental exploration and focused attention. This study shows that, under tightly controlled conditions, task-evoked pupil responses, an LC activity proxy, are lower in individuals with ASD than in controls, but only in the presence of task-irrelevant stimuli. This suggests that individuals with ASD evince atypical modulation of LC activity in accordance with changes in attentional demands, offering a mechanistic account for attentional atypicalities in ASD.