Charles Tijus

The mere perception of eye contact increases arousal during a word-spelling task

Abstract Eye contact is a highly salient and fundamentally social signal. This entails that the mere perception of direct gaze may trigger differentiated neurobehavioral responses as compared to other gaze directions. We investigated this issue using a visual word-spelling task where faces under different gaze directions and head orientations were displayed on-screen concomitantly with the words. We show evidence for automatic increase of skin conductance response (SCR), indicative of arousal, associated with the perception of direct gaze as compared to both averted gaze and closed eyes. Moreover, the perception of averted gaze was associated with an increase of electromyographic (EMG) corrugator activity. These effects were observed in two demanding word-spelling tasks, but not in a simple letter decision task. We propose to interpret these findings in terms of the social value of direct and averted gaze and conclude that some circumstances such as the task at hand may be essential for uncovering the neurobehavioral responses associated with the perception of others’ gaze.

Cognitive Processes in Artistic Creation: Toward the Realization of a Creative Machine

The author argues that Artificial Intelligence concepts will not lead to the development of a truly creative machine if limited merely to simulating the peripheral processes of the artist. After identifying five different creative processes discovered through collecting and evaluating data on the practices of 31 plastic artists, the author proposes a new paradigm for a creative machine, which involves the matching of schemata to establish new links between separate universes.

Appearance-based gaze tracking with spectral clustering and semi-supervised Gaussian process regression

Two of the challenges in appearance-based gaze tracking are: 1) prediction accuracy, 2) the efficiency of calibration process, which can be considered as the collection and analysis phase of labelled and unlabelled eye data. In this paper, we introduce an appearance-based gaze tracking model with a rapid calibration. First we propose to concatenate local eye appearance Center-Symmetric Local Binary Pattern(CS-LBP) descriptor for each subregion of eye image to form an eye appearance feature vector. The spectral clustering is then introduced to get the supervision information of eye manifolds on-line. Finally, taking advantage of eye manifold structure, a sparse semi-supervised Gaussian Process Regression(GPR) method is applied to estimate the subjects gaze coordinates. Experimental results demonstrate that our system with an efficient and accurate 5-points calibration not only can reduce the run-time cost but also can lead to a better accuracy result of 0.9°.

Zygomatic Smile Detection: The Semi-Supervised Haar Training of a Fast and Frugal System: A Gift to OpenCV Community

Five different OpenCV-compatible XML haarcascades of zygomatic smile detectors as well as five SMILEsamples from which these detectors were derived had been trained and are presented hereby as a new open source SMILEsmileD package. Samples have been extended in an incremental learning fashion, exploiting previously trained detector in order to add and label new elements of positive example set. After coupling with already known face detector, overall AUC performance ranges between 77%-90.5% when tested on JAFFE dataset and <1ms per frame speed of smile detection is achieved when tested on webcam-obtained videos. Observed results indicate that a semi-supervised incremental method which implements an existing haar-feature based classifier in order to to extend a sample from which a new classifier will be derived can be a method of 1) performance augmentation and 2) reduction of the amount of work dedicated to manual labeling of regions of interest in the positive example sample.

Chapter 2: The Design, Understanding and Usage of Pictograms

This chapter summarizes the results of studies investigating how to use pictograms to convey safety information in workplace, as well as some new insights into how to study the effects of iconic nature of pictogram information. First, it defines what a pictogram is and then describes empirical data reported in specialized domains publications, when assessing the role of pictograms in conveying recommended modes of behavior. The chapter discuses the theoretical contributions of semiotics, of cognitive psychology for pictogram readability and understanding, and of the contextual categorization approach, a theory that accounts for contextual effects involved in the understanding of pictograms. It describes the methodology, some examples and some outcomes from studies on creation of taxonomies and ontologies of pictograms for evaluation and design. Finally, it provides recommendations for ergonomic conception of pictograms, in order to improve the readability and understanding of pictograms used to convey user safety information in workplace. Keywords: contextual effects; design; ergonomic conception; ontologies; safety information; semiotics; taxonomies; understanding of pictograms

Immersive, Emotive and Cognitive effects of HDTV: Evaluating Effects of Image Definition on Presence

In the search of effects of image quality, we evaluated the effects of HDTV on Presence: Immersion, Emotion and Cognition. The set-up of the experiments involved respondents being wired up and recorded during two different viewing scenarios: in the first, they watched SD and HD versions of the same content being broadcast simultaneously on two identical adjacent HD-Ready flat-screen displays for ten minutes; in the second, they were asked to choose with a remote control and watch SD or HD programming from a selection offering content divided 50/50 between the two modes. In the first scenario, when the same content could be watched either in SD or HD over a ten-minute period, respondents spent 70% of the ten-minute period watching HD. 77% of them said they preferred HD; and 76% of them found that the quality of the HD picture was better. In the second scenario, when respondents were free to hop between different content modes as they saw fit, they spent 25% more time watching HD content before switching to a different channel. On average, they also chose HD programming 12.5% more than SD versions of the same programs. Physiological monitoring of the test subjects, combined with questionnaires, also found that HD both increased the legibility of, and the amount of ‘reading’ of the images, increasing comprehension of what was being shown, and stimulating viewers’ curiosity and imagination. Keywords-component; HDTV, Immersion, Emotion, Cognition

Counterfactual Based Innovation: A Galois Lattice Approach of Creative Thinking

Counterfactual reasoning is a form of modal reasoning about the way things could be (have been, would be) in contrast to reasoning about the way things actually are (were). This form of reasoning is elusive and provides reasoning bias. However, in the framework of a cognitive science theory that advocate that people reason using factual categories, counterfactuals being category counterfactuals, seen as same level collateral complementary categories of the factual category, provide a basis for innovation processes that differs from analogical thinking and case-based theory. The modeling shows how to implement tools for innovative solutions using Galois lattices and contextual graphs.

Explanation as contextual categorization

Our concern is the explanation generation in a representation based on contextual categorization. We point out that the explicit consideration of the context is necessary for the generation of relevant explanations. We present how the model captures the context necessary for explanation and we report some results compatible with the hypothesis that explanation is based on categorical networks according to a model based on the Galois lattice.

Applying Galois Lattices to the Interactions of the Virgin and Child in Bellini's Paintings

We discuss in our work a method for the semantic description and analysis of figure paintings [1]. Our method consists in the construction of a Galois lattice, which is a hierarchy of categories [2], and in the use of current theories of categorization in cognitive psychology and of mother-child interaction in developmental psychology. We have applied this method to a sample of 98 pictorial representations of the Virgin and Child attributed to Giovanni Bellini (1426–1516) [3]. Coding mother-infant interaction, we find Fogel’s concept of “frame” [4] appropriate for defining interaction units as co-regulated patterns in specific contexts: for instance, the social frame depicts mutual attachment behaviors and orientation of the subjects toward each other. For dyadic relations (within the dyad alone), we coded: (1) the category of attachment, related to affective sharing and proximityseeking (we distinguished three levels of attachment by considering mutuality of gaze, smiling, proximity and manual contact; (2) conflict of intention (for instance: autonomy of the Child versus dependence on the Virgin); and (3) scaffolding (the Virgin helps the Child to do what he cannot do by himself, for instance to keep upright or to walk). In the extra-dyadic situation (that of the dyad interacting with an object), we distinguished: (1) joint action; (2) offering; (3) request of attention (via the object); and (4) designation (or “proto-declarative intention”: the Child shows an object to his mother)[5]. Our analysis employed our computer program, Semantic Implication of