Reinforcement and learning

Evidence has been accumulating to support the process of reinforcement as a potential mechanism in speciation. In many species, mate choice decisions are influenced by cultural factors, including learned mating preferences (sexual imprinting) or learned mate attraction signals (e.g., bird song). It has been postulated that learning can have a strong impact on the likelihood of speciation and perhaps on the process of reinforcement, but no models have explicitly considered learning in a reinforcement context. We review the evidence that suggests that learning may be involved in speciation and reinforcement, and present a model of reinforcement via learned preferences. We show that not only can reinforcement occur when preferences are learned by imprinting, but that such preferences can maintain species differences easily in comparison with both autosomal and sex-linked genetically inherited preferences. We highlight the need for more explicit study of the connection between the behavioral process of learning and the evolutionary process of reinforcement in natural systems.     

Hydrological Forecasting in the Oder Estuary using a Three- Dimensional Hydrodynamic Model

A three-dimensional operational hydrodynamic model, developed at the Institute of Oceanography, University of Gdańsk was used to forecast hydrological conditions in the Oder Estuary. The model was based on the coastal ocean circulation model known as the Princeton Ocean Model (POM). Because of wind-driven water backup in the Oder mouth, a simplified operational model of river discharge, based on water budget in a stream channel, was developed. Linking these two models into a single system made it possible to forecast water levels, currents, water temperature, and salinity in the estuary. A good fit between the observed and computed data allowed to consider the model as a reliable environmental tool. Obtaining a hydrological forecast via a quick website access gives potential users an opportunity to predict the day-by-day course of processes that may affect different areas of human life and activities, e.g., navigation, port operations, flood protection of coastal areas; the predictions may also be used in studies of coastal processes in the estuary.     

Temporal species richness‐biomass relationships along successional gradients

Abstract. Diversity‐biomass relationships are frequently reported to be hump‐shaped over space at a given time. However, it is not yet clear how diversity and biomass change simultaneously and how they are related to each other over time (e.g. in succession) at one locality. This study develops a temporal model based on the projected changes of various community variables in a generalized terrestrial environment after fire and uses post‐fire succession data on Santa Monica Mountains of southern California and other published succession data to examine the temporal diversity‐biomass relationships. The results indicate that in the early stages of succession, both diversity and biomass increase and a positive relationship appears, while in the late stages of succession, biomass continued to increase but diversity usually declines; thus a negative relationship may be observed. When the scales of measurement become sufficiently large so that the measured diversity and biomass cross various stages of succession, a ‘hump‐shaped’ relationship can emerge. The diversity‐biomass relationship appears to be concordant in space and time when appropriate scales are used. Formerly proposed explanations for spatial patterns may well apply to the temporal patterns (particularly colonization, facilitation and competitive exclusion).     

An exploration of fixed and random effects selection for longitudinal binary outcomes in the presence of nonignorable dropout

We explore a Bayesian approach to selection of variables that represent fixed and random effects in modeling of longitudinal binary outcomes with missing data caused by dropouts. We show via analytic results for a simple example that nonignorable missing data lead to biased parameter estimates. This bias results in selection of wrong effects asymptotically, which we can confirm via simulations for more complex settings. By jointly modeling the longitudinal binary data with the dropout process that possibly leads to nonignorable missing data, we are able to correct the bias in estimation and selection. Mixture priors with a point mass at zero are used to facilitate variable selection. We illustrate the proposed approach using a clinical trial for acute ischemic stroke.     

Nanoparticle transport and delivery in a heterogeneous pulmonary vasculature

Quantitative understanding of nanoparticles delivery in a complex vascular networks is very challenging because it involves interplay of transport, hydrodynamic force, and multivalent interactions across different scales. Heterogeneous pulmonary network includes up to 16 generations of vessels in its arterial tree. Modeling the complete pulmonary vascular system in 3D is computationally unrealistic. To save computational cost, a model reconstructed from MRI scanned images is cut into an arbitrary pathway consisting of the upper 4-generations. The remaining generations are represented by an artificially rebuilt pathway. Physiological data such as branch information and connectivity matrix are used for geometry reconstruction. A lumped model is used to model the flow resistance of the branches that are cut off from the truncated pathway. Moreover, since the nanoparticle binding process is stochastic in nature, a binding probability function is used to simplify the carrier attachment and detachment processes. The stitched realistic and artificial geometries coupled with the lumped model at the unresolved outlets are used to resolve the flow field within the truncated arterial tree. Then, the biodistribution of 200 nm, 700 nm and 2 µm particles at different vessel generations is studied. At the end, 0.2–0.5% nanocarrier deposition is predicted during one time passage of drug carriers through pulmonary vascular tree. Our truncated approach enabled us to efficiently model hemodynamics and accordingly particle distribution in a complex 3D vasculature providing a simple, yet efficient predictive tool to study drug delivery at organ level.     

The oxo/peroxo debate: a nonheme iron perspective

The oxygen activation mechanisms proposed for nonheme iron systems generally follow the heme paradigm in invoking the involvement of iron-peroxo and iron-oxo species in their catalytic cycles. However, the nonheme ligand environments allow for end-on and side-on dioxygen coordination and impart greater flexibility in the modes of dioxygen activation. The currently available evidence for nonheme iron-peroxo and iron-oxo intermediates is summarized and discussed in light of the ongoing discussion on the nature of the oxidant(s) in heme enzymes.     

Partition of lindane in synthetic and native membranes

Partition coefficients of the insecticide γ-1,2,3,4,5,6-hexachlorocyclohexane (trivially, lindane) were determined in model and native membranes. Partition in egg phosphatidylcholine bilayers decreases linearly with temperature, over a range (10–40°C) at which the lipid is in the liquid-crystalline state. Addition of 50 mol% cholesterol dramatically decreases partition (2100 falls to 100, at 10°C) and abolishes the temperature dependence. First-order phase transitions of dimyristoyl-, dipalmitoyl- and distearoylphosphatidylcholines (DMPC, DPPC and DSPC) are accompanied by a sharp increase in lindane partition. Apparently, the insecticide is easily accommodated in bilayers of short-aliphatic-chain lipids, since the partitions were 2450, 600 and 50 in DMPC, DPPC and DSPC, respectively, at temperatures 10 Cdeg below the midpoint of their transitions. The lindane partition sequence in native membranes is as follows: mitochondria, sarcoplasmic reticulum, myelin, brain microsomes and erythrocytes. This sequence correlates reasonably well with the relative content of cholesterol and is similar in liposomes of total extracted lipids, although the absolute partitions showed decreased values. Therefore, the presence of proteins in native membranes contributes to the insecticide partition, probably by favouring its interaction with lipids.     

A bar model for the pump and channel function of the reconstituted Na+,K+-ATPase

Purified Na+,K+-ATPase is treated with trypsin. The altered enzyme is then reconstituted into liposomes and the change in active and passive Na+,K+-fluxes is recorded. Trypsin treatment transforms the slow passive Na+,K+-fluxes into leaks. The leak formation is correlated with the degree of proteolysis and the associated decrease in Na+,K+-ATPase activity. The active Na+,K+-transport capacity decreases in parallel with the passive transport. It is thus proposed that the Na+,K+-ATPase molecule primarily contains unspecific transmembrane tunnels that are rendered ion-selective by transverse bars of specific length (bar model).     

Vague-to-crisp dynamics of percept formation modeled as operant (selectionist) process

We model the vague-to-crisp dynamics of forming percepts in the brain by combining two methodologies: dynamic logic (DL) and operant learning process. Forming percepts upon the presentation of visual inputs is likened to model selection based on sampled evidence. Our framework utilizes the DL in selecting the correct “percept” among competing ones, but uses an intrinsic reward mechanism to allow stochastic online update in lieu of performing the optimization step of the DL framework. We discuss the connection of our framework with cognitive processing and the intentional neurodynamic cycle.