Craig B. Laramee

Building Trust in Public and Nonprofit Networks: Personal, Dyadic, and Third-Party Influences

This article provides greater understanding of factors influencing interpersonal trust in networks composed of public and nonprofit service providers. The present theoretical model identifies propensity to trust, the perceived trustworthiness of the trustee, the relationship between the trustee and trustor, and third-party relationships as influencing interpersonal trust. The model is tested using action research data collected from a network of local social service providers. Key findings include the following: (a) Successful past cooperation between a trustor and a trustee and structural equivalence increase the likelihood the trustor will perceive the trustee as trustworthy; (b) the frequency of interactions between the trustor and trustee, trust transferability, and the perceived trustworthiness of the trustee have a direct, positive impact on whether the trustor trusts the trustee; and (c) trust between the trustor and trustee has a positive impact on expected future cooperation.

Effect of Power-Frequency Magnetic Fields on Genome-Scale Gene Expression in Saccharomyces cerevisiae

Abstract Nakasono, S., Laramee, C., Saiki, H. and McLeod, K. J. Effect of Power-Frequency Magnetic Fields on Genome-Scale Gene Expression in Saccharomyces cerevisiae. Radiat. Res. 160, 25–37 (2003). To estimate the effect of 50 Hz magnetic-field exposure on genome-wide gene expression, the yeast Saccharomyces cerevisiae was used as a model for eukaryotes. 2D PAGE (about 1,000 spots) for protein and cDNA microarray (about 5,900 genes) analysis for mRNA were performed. The cells were exposed to 50 Hz vertical magnetic fields at 10, 150 or 300 mT r.m.s. for 24 h. As positive controls, the cells were exposed to aerobic conditions, heat (40°C) or minimal medium. The 2D PAGE and microarray analyses for the positive controls showed high-confidence differential expression of many genes including those for known or unknown proteins and mRNAs. For magnetic-field exposure, no high-confidence changes in expression were observed for proteins or genes that were related to heat-shock response, DNA repair, respiration, protein synthesis and the cell cycle. Principal component analysis showed no statistically significant difference in principal components, with only insignificant differences between the magnetic-field intensities studied. In contrast, the principal components for the positive controls were significantly different. The results indicate that a 50 Hz magnetic field below 300 mT did not act as a general stress factor like heat shock or DNA damage, as had been reported previously by others. This study failed to find a plausible differential gene expression that would point to a possible mechanism of an effect of magnetic fields. The findings provide no evidence that the magnetic-field exposure alters the fundamental mechanism of translation and transcription in eukaryotic cells.

Evolving spiking neural networks for robot control

Abstract We describe a sequence of experiments in which a robot “brain” was evolved to mimic the behaviours captured under control of a heuristic rule program (imitation learning). The task was light-seeking while avoiding obstacles using binocular light sensors and a trio of IR proximity sensors. The “brain” was a spiking neural network simulator whose parameters were tuned by a genetic algorithm, where fitness was assessed by the closeness to target output spike trains. Spike trains were frequency encoded. The network topology was manually designed, and then modified in response to observed difficulties during evolution. We noted that good performance seems best approached by judicious mixing of excitation and inhibition. Besides robotic applications, the domain of “smart” prosthetics also appears promising.

Generative Network Automata: A Generalized Framework for Modeling Adaptive Network Dynamics Using Graph Rewritings

A variety of modeling frameworks have been proposed and utilized in complex systems studies, including dynamical systems models that describe state transitions on a system of fixed topology, and self-organizing network models that describe topological transformations of a network with little attention paid to dynamical state changes. Earlier network models typically assumed that topological transformations are caused by exogenous factors, such as preferential attachment of new nodes and stochastic or targeted removal of existing nodes. However, many real-world complex systems exhibit both state transition and topology transformation simultaneously, and they evolve largely autonomously based on the system’s own states and topologies. Here we show that, by using the concept of graph rewriting, both state transitions and autonomous topology transformations of complex systems can be seamlessly integrated and represented in a unified computational framework. We call this novel modeling framework “Generative Network Automata (GNA)”. In this chapter, we introduce basic concepts of GNA, its working definition, its generality to represent other dynamical systems models, and some of our latest results of extensive computational experiments that exhaustively swept over possible rewriting rules of simple binary-state GNA. The results revealed several distinct types of the GNA dynamics.

Elevation of heat shock gene expression from static magnetic field exposure in vitro

Previously, we found that extremely low frequency (ELF) electric fields were able to elicit an approximate 3.5-fold increase in heat shock gene expression, a response which may have applicability to cancer therapy. Based on recent studies demonstrating the ability of magnetic fields to influence gene expression, we hypothesized that low level static magnetic fields may be able to affect heat shock gene expression while avoiding some of the clinical difficulties that arise with electric fields. Transfected rat primary cells in monolayer were exposed to magnetic fields of 1 to 440 mT for 16, 24, or 48 h starting at 24 and 48 h post transfection. Heat shock protein (HSP70) expression, as indicated by a promoter linked luciferase reporter, was followed for up to 96 h and showed a dependence on flux density, exposure duration, and start time post transfection. A nonlinear response was observed for increasing flux density with a maximum of a 3.5-fold increase in expression for 48 h of exposure starting 48 h after transfection. These results demonstrate an enhancement of gene expression similar in magnitude to that observed with external electric field exposure, while eliminating many of the clinical complications.

Evolving Spike Neural Network Sensors to Characterize the Alcoholic Brain Using Visually Evoked Response Potential

The electrical activity of the brain in response to a visual stimulus can be recorded using EEG. These signals are complex spatially-distributed time series. Here we investigate if it is possible to find hidden temporal patterns in these evoked electrical signals that could characterize the alcoholic brain. We have developed a technology for evolving spike neural network (SNN) sensors for detecting such hidden patterns in time-varying signals. The evolutionary computation involves a novel chromosome structure and a hybrid crossover operator for it. We have also developed a design rule for SNN-based temporal pattern detectors (TPD) that can detect a predefined inter-spike interval pattern in an incoming spike train. The design rule eliminates the need to tune the network parameters leaving only the design specifications to be learned. The primary goal of the evolutionary process is to select a set of EEG leads along with weights and to evolve the design specifications for the TPDs. After converting the composite EEG signal to a spike train, the TPDs are evaluated based on their ability to distinguish the alcoholic and the control cases. The early results suggest that this approach may be reliably used for characterizing the alcoholic brain.

Intelligent Alarm Processing into Clinical Knowledge

Alarmed physiological monitors have become a standard part of the ICU. While the alarms generated by these monitors can be important indicators of an altered physiological condition, most are unhelpful to medical staff due to a high incidence of false and clinically insignificant alarms. High numbers of false/insignificant alarms can lead to several adverse consequences such as increased patient anxiety,distraction of clinicians, and decreased efficiency in delivery of care. Furthermore, repeated false/insignificant alarms may increase the chance that healthcare providers ignore clinically significant alarms. In this paper we review the current state of intelligent alarm processing and describe an integrated systems methodology to extract clinically relevant information from physiological data. Such a method would aid significantly in the reduction of false alarms and provide nursing staff with a more reliable indicator of patient condition.

Induction of heat shock gene expression in RAT1 primary fibroblast cells by ELF electric fields.

Recent studies have demonstrated that the Ku70 gene fragment can be placed in the anti-sense orientation under the control of a heat-inducible heat shock protein 70 (HSP70) promoter and activated through heat shock exposure. This results in attenuation of the Ku70 protein expression, inhibiting cellular repair processes, and sensitizing the transfected cells to exposures such as the ionizing radiation exposures used clinically. However, achieving the tissue temperatures necessary to thermally induce the HSP70 response presents significant limitations to the clinical application of this strategy. Previous findings suggest an alternative approach to inducing a heat shock response, specifically through the use of extremely low frequency (ELF) electrical field stimulation. To further pursue this approach, we investigated HSP70 responses in transfected rat primary fibroblast (RAT1) cells exposed to 10 Hz electric fields at intensities of 20-500 V/m. We confirmed that low frequency electric fields can induce HSP70 heat shock expression, with peak responses obtained at 8 h following a 2 h field exposure. However, the approximate threefold increase in expression is substantially lower than that obtained using thermal stimulation, raising questions of the clinical utility of the response.

SSNNS -: a suite of tools to explore spiking neural networks

We are interested in engineering smart machines that enable backtracking of emergent behaviors. Our SSNNS simulator consists of hand-picked tools to explore spiking neural networks in more depth with flexibility. SSNNS is based on the Spike Response Model (SRM) with capabilities for short and long term memory. A genetic algorithm, namely CHC, is used independently to generate such example systems that produce patterns of interest. Foundational work in the growing field of spiking neural networks has shown that precise spike timing may be biologically more plausible and computationally powerful than traditional rate-based models[4][7]. We have been using evolution to discover neural configurations that produce patterns of interest.

A novel approach to signal classification with an application to identifying the alcoholic brain

Graphical abstractDisplay Omitted HighlightsEvolving temporal pattern detectors for signal classification.We make no assumptions regarding the spectral characteristics of the data.The classification accuracies are comparable to the conventional techniques.Located EEG-sensors that showed abnormal electrical behavior for the alcoholics.Evolutionary learning paradigm unified feature extraction and classification steps. We introduce a novel approach to signal classification based on evolving temporal pattern detectors (TPDs) that can find the occurrences of embedded temporal structures in discrete time signals and illustrate its application to characterizing the alcoholic brain using visually evoked response potentials. In contrast to conventional techniques used for most signal classification tasks, this approach unifies the feature extraction and classification steps. It makes no prior assumptions regarding the spectral characteristics of the data; it merely assumes that some temporal patterns exist that distinguish two classes of signals and therefore could be applied to new signal classification tasks where a body of prior work identifying important features does not exist. Evolutionary computation (EC) discovers a classifier by simply learning from the time series samples.The alcoholic classification (AC) problem consists of 2 sub-tasks, one spatial and one temporal: choosing a subset of electroencephalogram leads used to create a composite signal (the spatial task), and detecting temporal patterns in this signal that are more prevalent in the alcoholics than the controls (the temporal task). To accomplish this, a novel representation and crossover operator were devised that enable multiple feature subset tasks to be solved concurrently. Three TPD techniques are presented that differ in the mechanism by which partial credit is assigned to temporal patterns that deviate from the specified pattern. An EC approach is used for evolving a subset of sensors and the TPD specifications. We found evidence that partial credit does help evolutionary discovery. Regions on the skull of an alcoholic subject that produced abnormal electrical activity compared to the controls were located. These regions were consistent with prior findings in the literature. The classification accuracy was measured as the area under the receiver operator characteristic curve (ROC); the ROC area for the training set varied from 90.32% to 98.83% and for the testing set it varied from 87.17% to 95.9%.