Gaze and informativeness during category learning: Evidence for an inverse relation

In what follows, we explore the general relationship between eye gaze during a category learning task and the information conveyed by each member of the learned category. To understand the nature of this relationship empirically, we used eye tracking during a novel object classification paradigm. Results suggest that the average fixation time per object during learning is inversely proportional to the amount of information that object conveys about its category. This inverse relationship may seem counterintuitive; however, objects that have a high-information value are inherently more representative of their category. Therefore, their generality captures the essence of the category structure relative to less representative objects. As such, it takes relatively less time to process these objects than their less informative companions. We use a general information measure referred to as representational information theory (Vigo, 2011a, 2013a) to articulate and interpret the results from our experiment and compare its predictions to those of three models of prototypicality.     

Developing race categories in infancy via Bayesian face recognition

The other-race effect emerges during infancy following the perceptual narrowing of face recognition. Other-race faces that were previously discriminable in early infancy cannot be distinguished by older infants. I discuss a Bayesian model of this process that posits that the other-race effect may be a consequence of learning to distinguish between intrapersonal variation (changes to face appearance that preserve identity) and extrapersonal variation (changes that do not preserve identity) in a visual environment in which a subset of race categories dominate. I demonstrate that race categories, which I have previously argued are a critical precursor to the emergence of the other-race effect in infancy, are a natural by-product of this model. Perceptual narrowing for race may thus be a natural consequence of visual experience and the estimation of face variability based on a growing number of exemplars. I describe the basic architecture of the model, its applicability to a range of visual learning scenarios, and identify critical choices one faces in applying the model to a specific perceptual task. Despite the success of the model in accounting for these behavioural results, I conclude by identifying important shortcomings of the model and describe important challenges for future efforts to characterize the development of the other-race effect computationally.     

A biased-competition approach to spatial cueing: Combining empirical studies and computational modelling

We examined the effects of cue luminance on visual orienting. Experiment 1 established that the commonly-found early facilitation and late inhibition of return (IOR) effects were independent of cue luminance with single cues in terms of their amplitude, although IOR was delayed in the low compared to the high luminance cue condition. In contrast, Experiment 2 revealed that, with dual cues of mixed luminance, both facilitation and IOR effects were found only with bright cues. When cues had equal luminance, however, there were cueing effects for two cued locations but only when the cues were bright. The data were accommodated in a neural network model of biased competition in which cueing effects emerge at more than one location provided input activation is sufficient to overcome competitive damping of the selection system.     

Occupational well-being as a mediator between job insecurity and turnover intention: Findings at the individual and work department levels

This study examined the relationship between job insecurity and turnover intention by applying occupational well-being (exhaustion, vigour) as a mediator. The study was inspired by two theories: the conservation of resources and emotional contagion theories. We investigated the relationships at the individual and work department levels by utilizing Multi-Level Structural Equation Modeling (ML-SEM) with the aim of clarifying whether the mediating mechanism was similar at both levels. In addition, we examined the relationships across the levels (cross-level interactions). Self-report data for the study were obtained from Finnish University staff (N = 2137 individual respondents from 78 work departments). The analyses resulted in three main findings. First, job insecurity, turnover intention, and occupational well-being were found, to some extent (2–6%), to be shared experiences within work departments. Second, we found that low occupational well-being (high exhaustion, low vigour) partly mediated the relationship between job insecurity and turnover intention at both levels of analysis. Third, the results on cross-level interactions revealed that the lower the level of well-being at the work department level, the stronger the negative effect of job insecurity on well-being at the individual level. Thus, if poor well-being characterizes the work department, this may strengthen the negative relationship between job insecurity and well-being at the individual level.     

Stimulus type and the list strength paradigm

In recognition memory, increasing the strength of studied items does not reduce performance on other items, an effect dubbed the null list strength effect (LSE). While this finding has been replicated many times, it has rarely been tested using stimuli other than single words. Kinnell and Dennis (2012 Kinnell, A., & Dennis, S. (2012). The role of stimulus type in list length effects in recognition memory. Memory & Cognition, 40, 311–325. doi: 10.3758/s13421-011-0164-2[Crossref], [PubMed], [Web of Science ®] , [Google Scholar]) recently tested for the presence of list length effects using non-word stimulus classes while controlling for the confounds that are present in list length designs. Small list length effects were found for fractal and face images. We adopted the same paradigm and stimuli used by Kinnell and Dennis to test whether these stimuli would be susceptible to list strength effects as well. We found significant LSEs for fractal images, but null LSEs for face images and natural scene photographs. Stimuli other than words do appear to be susceptible to list strength effects, but these effects are small and restricted to particular stimulus classes, as is the case in list length designs. Models of memory may be able to address differences between these stimulus classes by attributing differences in representational overlap between the stimulus classes.     

Grounding Bayesian accounts of numerosity and variability effects in a similarity-based framework: the case of self-organising maps

ABSTRACT

In this paper, we propose self-organising maps as possible candidates to explain the psychological mechanisms underlying category generalisation. Self-organising maps are psychologically and biologically plausible neural network models that can learn after limited exposure to positive category examples, without any need of contrastive information. They reproduce human behaviour in category generalisation, in particular, the Numerosity and Variability effects, which are usually explained with Bayesian tools. Where category generalisation is concerned, self-organising maps deserve attention to bridge the gap between the computational level of analysis in Marr's hierarchy (where Bayesian models are often situated) and the algorithmic level of analysis in which plausible mechanisms are described.     

Heuristic Methods for Computer Ethics

The domain of "procedural ethics" is the set of reflective and deliberative methods that maximize the reliability of moral judgment. While no general algorithmic method exists that will guarantee the validity of ethical deliberation, non-algorithmic "heuristic" methods can guide and inform the process, making it significantly more robust and dependable. This essay examines various representative heuristic procedures commonly recommended for use in applied ethics, maps them into a uniform set of twelve stages, identifies common faults, then shows how the resulting stage-by-stage decision-making model could be adapted for general use and for use in computer ethics.     

Lilliputian Computer Ethics

This essay considers some ethical issues of nanotechnology and quantum computing, particularly the issue of privacy, and questions related to artificial intelligence, implants, and virtual reality. It then examines the claim that research in this field should be halted.