The event-related brain potential (ERP) literature concerning the impact of emotional content on visual word processing is reviewed and related to general knowledge on semantics in word processing: emotional connotation can enhance cortical responses at all stages of visual word processing following the assembly of visual word form (up to 200 ms), such as semantic access (around 200 ms), allocation of attentional resources (around 300 ms), contextual analysis (around 400 ms), and sustained processing and memory encoding (around 500 ms). Even earlier effects have occasionally been reported with subliminal or perceptual threshold presentation, particularly in clinical populations. Here, the underlying mechanisms are likely to diverge from the ones operational in standard natural reading. The variability in timing of the effects can be accounted for by dynamically changing lexical representations that can be activated as required by the subjects' motivational state, the task at hand, and additional contextual factors. Throughout, subcortical structures such as the amygdala are likely to contribute these enhancements. Further research will establish whether or when emotional arousal, valence, or additional emotional properties drive the observed effects and how experimental factors interact with these. Meticulous control of other word properties known to affect ERPs in visual word processing, such as word class, length, frequency, and concreteness and the use of more standardized EEG procedures is vital. Mapping the interplay between cortical and subcortical mechanisms that give rise to amplified cortical responses to emotional words will be of highest priority for future research.

Emotional and semantic networks in visual word processing: insights from ERP studies.

Recent studies show that introducing sound from water features in urban open spaces may reduce the loudness of road traffic noise, but it is not clear in which situations this measure also improves overall soundscape quality. This work describes a listening experiment on loudness, pleasantness, and eventfulness of stimuli that combine road traffic noise with fountain or bird sound at different sound levels. Adding fountain sound reduced the loudness of road traffic noise only if the latter had low temporal variability. Conversely, adding bird sound significantly enhanced soundscape pleasantness and eventfulness, more than what was achieved by adding fountain sound.

/pdf/effects-of-natural-sounds-on-the-perception-of-road-traffic-8g62z55axy.pdf

Effects of natural sounds on the perception of road traffic noise

Thurstonian and Bradley-Terry models are the most commonly applied models in the analysis of paired comparison data. Since their introduction, numerous developments have been proposed in different areas. This paper provides an updated overview of these extensions, including how to account for object- and subject-specific covariates and how to deal with ordinal paired comparison data. Special emphasis is given to models for dependent comparisons. Although these models are more realistic, their use is complicated by numerical difficulties. We therefore concentrate on implementation issues. In particular, a pairwise likelihood approach is explored for models for dependent paired comparison data, and a simulation study is carried out to compare the performance of maximum pairwise likelihood with other limited information estimation methods. The methodology is illustrated throughout using a real data set about university paired comparisons performed by students.

Models for Paired Comparison Data: A Review with Emphasis on Dependent Data

As technical realization of highly and fully automated vehicles draws closer, attention is being shifted from sheer feasibility to the question of how an acceptable driving style and thus comfort can be implemented. It is increasingly important to determine, how highly automated vehicles should drive to ensure driving comfort for the now passive drivers. Thus far, only little research has been conducted to examine this issue. In order to lay a basis on how automated vehicles should drive to ensure passenger comfort, different variations of three central maneuvers were rated and analyzed. A simulator study (N = 72) was conducted in order to identify comfortable driving strategies. Three variations of lane changes, accelerations and decelerations were configured by manipulating acceleration and jerk, and thus the course of each maneuver. Furthermore, the influence of personality traits and self-reported driving style on preferences of differently executed automated maneuvers was analyzed. Results suggest keeping acceleration and jerk as small as possible for acceleration maneuvers. For lane changes, both small accelerations as well as an early motion feedback are advisable. Interestingly, decelerating as a manual driver would is rejected compared to two artificial alternatives. Moreover, no influence of personality traits on maneuver preference was found. Only self-reported driving style had a marginal effect on participants’ preferences. In conclusion, a recommendation for an automated driving style can be given, which was perceived as comfortable by participants regardless of their personality.

Comfort in automated driving: An analysis of preferences for different automated driving styles and their dependence on personality traits

Priors for Bayesian nonparametric latent feature models were originally developed a little over five years ago, sparking interest in a new type of Bayesian nonparametric model. Since then, there have been three main areas of research for people interested in these priors: extensions/generalizations of the priors, inference algorithms, and applications. This dissertation summarizes our work advancing the state of the art in all three of these areas. In the first area, we present a non-exchangeable framework for generalizing and extending the original priors, allowing more prior knowledge to be used in nonparametric priors. Within this framework, we introduce four concrete generalizations that are applicable when we have prior knowledge about object relationships that can be captured either via a tree or chain. We discuss how to develop and derive these priors as well as how to perform posterior inference in models using them. In the area of inference algorithms, we present the first variational approximation for one class of these priors, demonstrating in what regimes they might be preferred over more traditional MCMC approaches. Finally, we present an application of basic nonparametric latent features models to link prediction as well as applications of our non-exchangeable priors to tree-structured choice models and human genomic data.

/pdf/bayesian-nonparametric-latent-feature-models-27y0rm65an.pdf

Bayesian nonparametric latent feature models

Tversky (1972) has proposed a family of models for paired-comparison data that generalize the Bradley-Terry-Luce (BTL) model and can, therefore, apply to a diversity of situations in which the BTL model is doomed to fail. In this article, we present a Matlab function that makes it easy to specify any of these general models (EBA, Pretree, or BTL) and to estimate their parameters. The program eliminates the time-consuming task of constructing the likelihood function by hand for every single model. The usage of the program is illustrated by several examples. Features of the algorithm are outlined. The purpose of this article is to facilitate the use of probabilistic choice models in the analysis of data resulting from paired comparisons.

/pdf/a-matlab-function-to-estimate-choice-model-parameters-from-4l3xxo8i1p.pdf

A Matlab function to estimate choice model parameters from paired-comparison data

Paired comparisons require nothing more from a subject than a choice between two stimuli with respect to a specified criterion The subsequent analysis of such data by probabilistic choice models (PCMs) yields further advantages: ratio-scale measures of the stimuli can be derived, standard statistical theory can be employed for estimation and testing, and a theory of human decision making is incorporated in the scaling procedure

/pdf/probabilistic-choice-models-for-psychological-scaling-evcmps8l4x.pdf

Probabilistic Choice Models for Psychological Scaling

Despite extensive theoretical work on biologically plausible learning rules, clear evidence about whether and how such rules are implemented in the brain has been difficult to obtain. We consider biologically plausible supervised- and reinforcement-learning rules and ask whether changes in network activity during learning can be used to determine which learning rule is being used. Supervised learning requires a credit-assignment model estimating the mapping from neural activity to behavior, and, in a biological organism, this model will inevitably be an imperfect approximation of the ideal mapping, leading to a bias in the direction of the weight updates relative to the true gradient. Reinforcement learning, on the other hand, requires no credit-assignment model and tends to make weight updates following the true gradient direction. We derive a metric to distinguish between learning rules by observing changes in the network activity during learning, given that the mapping from brain to behavior is known by the experimenter. Because brain-machine interface (BMI) experiments allow for precise knowledge of this mapping, we model a cursor-control BMI task using recurrent neural networks, showing that learning rules can be distinguished in simulated experiments using only observations that a neuroscience experimenter would plausibly have access to.

Distinguishing Learning Rules with Brain Machine Interfaces

Neurons in the brain face the challenge of representing sensory stimuli in a way that accurately encodes the features of these stimuli while minimizing the eﬀects of noise. This thesis will use the concept of mutual information from information theory, which quantiﬁes the amount of information one variable can tell us about another and vice versa, to better understand neural coding in the auditory cortex. Previous research has been done in maximizing mutual information to better understand neural behavior patterns in the visual cortex, with limited auditory ﬁndings. We perform numerical optimization in Python to maximize information that a population of neurons contains about an auditory stimulus within the framework of information theory. This is done by ﬁrst ﬁnding the optimal width and location of tuning curves that characterize neural response to one dimensional stimuli (sound frequency), then updating the optimization algorithm to ﬁt two-dimensional stimuli (sound frequency and intensity). By testing the algorithm with a set of natural sound data, our computations show that in the latter case, optimal stimulus information is represented by a speciﬁc homogeneous population with similar response properties. Our ﬁndings provide a method to better understand neural representation in the auditory cortex, speciﬁcally, the relationship between neural response and natural sound stimuli.

Using Information Theory to Understand Neural Representation in the Auditory Cortex Hannah

Learning to perform a perceptual decision task is generally achieved through sessions of effortful practice with feedback. Here, we investigated how passive exposure to relevant stimuli, which is relatively effortless and does not require feedback, influences active learning. First, we trained mice in a sound-categorization task with various schedules combining passive exposure and active training. Mice that received passive exposure exhibited faster learning, regardless of whether this exposure occurred entirely before active training or was interleaved between active sessions. We next trained neural-network models with different architectures and learning rules to perform the task. Networks that use the statistical properties of stimuli to enhance separability of the data via unsupervised learning during passive exposure provided the best account of the behavioral observations. We further found that, during interleaved schedules, there is an increased alignment between weight updates from passive exposure and active training, such that a few interleaved sessions can be as effective as schedules with long periods of passive exposure before active training, consistent with our behavioral observations. These results provide key insights for the design of efficient training schedules that combine active learning and passive exposure in both natural and artificial systems.

/pdf/passive-exposure-to-task-relevant-stimuli-enhances-3odqz16e.pdf

Christian Schmid

Papers

A Matlab function to estimate choice model parameters from paired-comparison data

Probabilistic Choice Models for Psychological Scaling

Distinguishing Learning Rules with Brain Machine Interfaces

Using Information Theory to Understand Neural Representation in the Auditory Cortex Hannah

Passive exposure to task-relevant stimuli enhances categorization learning