What Is Social about Autism? The Role of Allostasis-Driven Learning

Scientific research on neuro-cognitive mechanisms of autism often focuses on circuits that support social functioning. However, autism is a heterogeneous developmental variation in multiple domains, including social communication, but also language, cognition, and sensory-motor control. This suggests that the underlying mechanisms of autism share a domain-general foundation that impacts all of these processes. In this Perspective Review, we propose that autism is not a social deficit that results from an atypical “social brain”. Instead, typical social development relies on learning. In social animals, infants depend on their caregivers for survival, which makes social information vitally salient. The infant must learn to socially interact in order to survive and develop, and the most prominent learning in early life is crafted by social interactions. Therefore, the most prominent outcome of a learning variation is atypical social development. To support the hypothesis that autism results from a variation in learning, we first review evidence from neuroscience and developmental science, demonstrating that typical social development depends on two domain-general processes that determine learning: (a) motivation, guided by allostatic regulation of the internal milieu; and (b) multi-modal associations, determined by the statistical regularities of the external milieu. These two processes are basic ingredients of typical development because they determine allostasis-driven learning of the social environment. We then review evidence showing that allostasis and learning are affected among individuals with autism, both neurally and behaviorally. We conclude by proposing a novel domain-general framework that emphasizes allostasis-driven learning as a key process underlying autism. Guided by allostasis, humans learn to become social, therefore, the atypical social profile seen in autism can reflect a domain-general variation in allostasis-driven learning. This domain-general view raises novel research questions in both basic and clinical research and points to targets for clinical intervention that can lower the age of diagnosis and improve the well-being of individuals with autism.

Details (Don't) Matter: Isolating Cluster Information in Deep Embedded Spaces

Deep clustering techniques combine representation learning with clustering objectives to improve their performance. Among existing deep clustering techniques, autoencoder-based methods are the most prevalent ones. While they achieve promising clustering results, they suffer from an inherent conflict between preserving details, as expressed by the reconstruction loss, and finding similar groups by ignoring details, as expressed by the clustering loss. This conflict leads to brittle training procedures, dependence on trade-off hyperparameters and less interpretable results. We propose our framework, ACe/DeC, that is compatible with Autoencoder Centroid based Deep Clustering methods and automatically learns a latent representation consisting of two separate spaces. The clustering space captures all cluster-specific information and the shared space explains general variation in the data. This separation resolves the above mentioned conflict and allows our method to learn both detailed reconstructions and cluster specific abstractions. We evaluate our framework with extensive experiments to show several benefits: (1) cluster performance – on various data sets we outperform relevant baselines; (2) no hyperparameter tuning – this improved performance is achieved without introducing new clustering specific hyperparameters; (3) interpretability – isolating the cluster specific information in a separate space is advantageous for data exploration and interpreting the clustering results; and (4) dimensionality of the embedded space – we automatically learn a low dimensional space for clustering. Our ACe/DeC framework isolates cluster information, increases stability and interpretability, while improving cluster performance.

Effects of sea birds and soil development on plant and soil nutritional parameters after 50 years of succession on Surtsey

Surtsey, the island that rose from the sea in a submarine eruption during 1963 to 1967, has been the subject of many studies on primary succession. These studies have intensified after the establishment of a seagull colony on the island in 1986. This paper reports on the results of a short sampling expedition in 2013 that intended to characterize the interactive effects of the seagull colony and of soil development on soil nutritional characteristics in the tephra sands that cover the underlying lava, as well as in plants growing inside and outside the seagull colony. Feces and pellets of the gulls were extremely rich in both nitrogen (N) and phosphorus (P) and δ15N analyses showed that N was transferred from pellets and feces to the tephra soils and subsequently taken up by the plants. The tephra soils not affected by the birds showed a high concentration of P compared to N. The concentration of both nutrients was much lower than in the soils of the bird colony. In general, variation in tephra soil depth had little effect on nutritional characteristics, except for the very low N concentration in deep soils. Thus, our results confirm the overriding effect of the seagull colony on Surtsey on nutritional characteristics of the developing soils and vegetation. Due to the very high P availability of the volcanic soils in combination with the high P input by the birds, vegetation productivity is N limited, despite the extremely high N input of 47 kg N ha-1 yr-1 that the birds add to the system. Our findings emphasize the extreme importance of bird colonies on the nutritional ecology of young, N-poor ecosystems.

Variations in Naturalistic Driving Behavior and Visual Perception at the Entrances of Short, Medium, and Long Tunnels

Driver behavior and visual perception are very important factors in the management of traffic accident risk at tunnel entrances. This study was undertaken to analyze the differences in driving behavior and visual perception at the entrances of three types of tunnels, namely, short, medium-length, and long tunnels, under naturalistic driving conditions. Using three driving behavior indicators (speed, deceleration, and position) and two visual perception indicators (fixation and saccade), the driving performance of twenty drivers at six tunnels (two tunnels per condition) was comparatively analyzed. The results revealed that the speed maintained by the drivers prior to deceleration with braking under the short-tunnel condition was significantly larger than that under the medium- and long-tunnel conditions and that the drivers had a greater average and maximum deceleration rates under the short-tunnel condition. A similar general variation of driver visual perception appeared under the respective tunnel conditions, with the number of fixations gradually increasing and the maximum saccade amplitude gradually decreasing as the drivers approached the tunnel portal. However, the variation occurred approximately 60 m earlier under the short-tunnel condition than under the medium- and long-tunnel conditions. Interactive correlations between driving behavior and visual perception under the three conditions were established. The commencement of active deceleration was significantly associated (with correlation factors of 0.80, 0.77, and 0.79 under short-, medium-, and long-tunnel conditions, respectively) with the point at which the driver saccade amplitude fell below 10 degrees for more than 3 s. The results of this study add to the sum of knowledge of differential driver performance at the entrances of tunnels of different lengths.

Modal Homotopy Type Theory

In[KF1] 1914, in an essay entitled ‘Logic as the Essence of Philosophy’, Bertrand Russell promised to revolutionize philosophy by introducing there the ‘new logic’ of Frege and Peano: “The old logic put thought in fetters, while the new logic gives it wings.” A century later, this book proposes a comparable revolution with a newly emerging logic, modal homotopy type theory. Russell’s prediction turned out to be accurate. Frege’s first-order logic, along with its extension to modal logic, is to be found throughout anglophone analytic philosophy. This book provides a considerable array of evidence for the claim that philosophers working in metaphysics, as well as those treating language, logic or mathematics, would be much better served with the new ‘new logic’. It offers an introduction to this new logic, thoroughly motivated by intuitive explanations of the need for all of its component parts—the discipline of a type theory, the flexibility of type dependency, the more refined homotopic notion of identity and a powerful range of modalities. Innovative applications of the calculus are given, including analysis of the distinction between objects and events, an intrinsic treatment of structure and a conception of modality both as a form of general variation and as allowing constructions in modern geometry. In this way, we see how varied are the applications of this powerful new language—modal homotopy type theory.

Ultra-broadband Optical Gain Engineering in Solution-processed QD-SOA Based on Superimposed Quantum Structure

In this work, the optical gain engineering of an ultra-broadband InGaAs/AlAs solution-processed quantum dot (QD) semiconductor optical amplifier using superimposed quantum structure is investigated. The basic unit in the proposed structure (QDs) is designed and fabricated using solution-processed methods with considerable cost-effectiveness, fabrication ease, and QDs size tunability up to various limits (0.1 nm up to the desired values), considering suitable synthesis methods. Increasing the number of QDs, the device can span more than 1.02 μm (O, C, S, and L bands) using only one type of material for all QDs, and is not restricted to this limit in case of using more QD groups. Also, it can manipulate the optical gain peak value, spectral coverage, and resonant energy for customized optical windows, among which 1.31 μm and 1.55 μm are simulated as widely-applicable cases for model validation. This makes the device a prominent candidate for ultra-wide-bandwidth and also customized-gain applications in general. Variation impact of homogeneous and inhomogeneous broadenings, injection current and number of QD groups on optical gain are explained in detail. Besides proposing a design procedure for implementation of an ultra-broadband optical gain using superimposed QDs in solution-processed technology, the proposed gain engineering idea using this technology provides practically infinite bandwidth and an easy way to realize. By introducing this idea, one more step is actually taken to approach the effectiveness of solution process technology.

Environmental conditions, not sugar export efficiency, limit the length of conifer leaves

Most conifer species have needle-shaped leaves that are only a few centimeters long. In general, variation in leaf size has been associated with environmental factors, such as cold or drought stress. However, it has recently been proposed that sugar export efficiency is the limiting factor for conifer needle length, based on the results obtained using a mathematical model of phloem transport. Here, phloem transport rates in long conifer needles were experimentally determined to test if the mathematical model accurately represents phloem transport. The validity of the model’s assumptions was tested by anatomical analyses and sugar quantification. Furthermore, various environmental and physiological factors were tested for their correlation with needle length. The results indicate that needle length is not limited by sugar transport efficiency, but, instead, by winter temperatures and light availability. The identification of factors that influence needle size is instrumental for using this trait as a variable in breeding programs.

Variational Histogram Equalization for Single Color Image Defogging

Foggy images taken in the bad weather inevitably suffer from contrast loss and color distortion. Existing defogging methods merely resort to digging out an accurate scene transmission in ignorance of their unpleasing distortion and high complexity. Different from previous works, we propose a simple but powerful method based on histogram equalization and the physical degradation model. By revising two constraints in a variational histogram equalization framework, the intensity component of a fog-free image can be estimated in HSI color space, since the airlight is inferred through a color attenuation prior in advance. To cut down the time consumption, a general variation filter is proposed to obtain a numerical solution from the revised framework. After getting the estimated intensity component, it is easy to infer the saturation component from the physical degradation model in saturation channel. Accordingly, the fog-free image can be restored with the estimated intensity and saturation components. In the end, the proposed method is tested on several foggy images and assessed by two no-reference indexes. Experimental results reveal that our method is relatively superior to three groups of relevant and state-of-the-art defogging methods.

On the Role of Mesoentropy in Thermodynamics of First Order Phase Transformations in Alloys

A thermodynamics of a new phase formation in a binary alloy, taking into account the chemical depletion and the size distribution of new phase nuclei giving the energy input at high temperatures, is considered. It is shown the principal possibility of obtaining of the size distribution function for nuclei in the framework of a general variation procedure for the Gibbs free energy functional. The ‘chemical depletion’ related to the difference of compositions of the new phase embryo and the solution of the solid metastable alloy is discussed. The role of the ‘mesoentropy’ for the determination of whole energy of first order phase transformation is manifested.