Biology subjects

Burnham, D.

Publications and source records attributed to Burnham, D..

3 recordsLinked to original sources

bioRxiv2022

Atypical cortical encoding of speech identifies children with Dyslexia versus Developmental Language Disorder

Slow cortical oscillations play a crucial role in processing the speech envelope, which is perceived atypically by children with Developmental Language Disorder (DLD) and developmental dyslexia. Here we use electroencephalography (EEG) and natural speech listening paradigms to identify neural processing patterns that characterize dyslexic versus DLD children. Using a story listening paradigm, we show that atypical power dynamics and phase-amplitude coupling between delta and theta oscillations characterize dyslexic and DLD children groups, respectively. We further identify EEG common spatial patterns (CSP) during speech listening across delta, theta and beta oscillations describing dyslexic versus DLD children. A linear classifier using four deltaband CSP variables predicted dyslexia status (0.77 AUC). Crucially, these spatial patterns also identified children with dyslexia in a rhythmic syllable task EEG, suggesting a core developmental deficit in neural processing of speech rhythm. These findings suggest that there are distinct atypical neurocognitive mechanisms underlying dyslexia and DLD.

neuroscience↗

bioRxiv2022

40 new specimens of Ichthyornis provide unprecedented insight into the postcranial morphology of crownward stem group birds.

Ichthyornis has long been recognized as a pivotally important fossil taxon for understanding the latest stages of the dinosaur-bird transition, but little significant new postcranial material has been brought to light since initial descriptions of partial skeletons in the 19th Century. Here, we present new information on the postcranial morphology of Ichthyornis from 40 previously undescribed specimens, providing the most detailed morphological assessment of Ichthyornis to date. The new material includes four partially complete skeletons and numerous well-preserved isolated elements, enabling new anatomical observations such as muscle attachments previously undescribed for Mesozoic euornitheans. Among the elements that were previously unknown or poorly represented for Ichthyornis, the new specimens include an almost-complete axial series, a hypocleideum-bearing furcula, radial carpal bones, fibulae, a complete tarsometatarsus bearing a rudimentary hypotarsus, and one of the first-known nearly complete three-dimensional sterna from a Mesozoic avialan. Several pedal phalanges are preserved, revealing a remarkably enlarged pes presumably related to foot-propelled swimming. Although diagnosable as Ichthyornis, the new specimens exhibit a substantial degree of morphological variation, some of which may relate to ontogenetic changes. Phylogenetic analyses incorporating our new data and employing alternative morphological datasets recover Ichthyornis stemward of Hesperornithes and Iaceornis, in line with some recent hypotheses regarding the topology of the crownward-most portion of the avian stem group, and we establish phylogenetically-defined clade names for relevant avialan subclades to help facilitate consistent discourse in future work. The new information provided by these specimens improves our understanding of morphological evolution among the crownward-most non-neornithine avialans immediately preceding the origin of crown group birds.

paleontology↗

bioRxiv2021

Learning to Predict in Networks with Heterogeneous and Dynamic Synapses

AO_SCPLOWBSTRACTC_SCPLOWA salient difference between artificial and biological neural networks is the complexity and diversity of individual units in the latter (Tasic et al., 2018). This remarkable diversity is present in the cellular and synaptic dynamics. In this study we focus on the role in learning of one such dynamical mechanism missing from most artificial neural network models, short-term synaptic plasticity (STSP). Biological synapses have dynamics over at least two time scales: a long time scale, which maps well to synaptic changes in artificial neural networks during learning, and the short time scale of STSP, which is typically ignored. Recent studies have shown the utility of such short-term dynamics in a variety of tasks (Masse et al., 2019; Perez-Nieves et al., 2021), and networks trained with such synapses have been shown to better match recorded neuronal activity and animal behavior (Hu et al., 2020). Here, we allow the timescale of STSP in individual neurons to be learned, simultaneously with standard learning of overall synaptic weights. We study learning performance on two predictive tasks, a simple dynamical system and a more complex MNIST pixel sequence. When the number of computational units is similar to the task dimensionality, RNNs with STSP outperform standard RNN and LSTM models. A potential explanation for this improvement is the encoding of activity history in the short-term synaptic dynamics, a biological form of long short-term memory. Beyond a role for synaptic dynamics themselves, we find a reason and a role for their diversity: learned synaptic time constants become heterogeneous across training and contribute to improved prediction performance in feedforward architectures. These results demonstrate how biologically motivated neural dynamics improve performance on the fundamental task of predicting future inputs with limited computational resources, and how learning such predictions drives neural dynamics towards the diversity found in biological brains.

neuroscience↗