This investigation examined whether children can code the relative proximity of two objects to a landmark and whether they use verbal or nonverbal strategies to remember a target location. Two- to 2.5-year-olds completed a memory task where they watched an experimenter hide two different toys in two identical containers placed 2 and 12 inches from a landmark. The experimenter either used neutral language (e.g., “here”; Experiment 1) or spatial labels (e.g., “close/far”; Experiment 2) to describe objects’ hiding locations. After hiding, children were carried outside the enclosure to a new viewpoint during a 10-second delay and then looked for a target toy. Experiment 2 also included language measures: parent reports of children’s general and relational vocabularies and performance on a language task, which measured children’s understanding of spatial (close/far) and color (red/blue) terms. We found that children successfully coded relative to proximity to a landmark in the memory task. However, hearing spatial labels during hiding in Experiment 2 did not improve performance relative to Experiment 1, and children’s spatial term comprehension in the language task did not predict memory task performance. We also found that children’s productive relational vocabulary predicted memory task performance; however, children’s color term comprehension in the language task was the strongest overall predictor of memory task performance. Collectively, these results suggest that children initially rely on a nonverbal strategy when coding relative proximity to a landmark in a memory task and that children who are better at forming abstract categories may code relative proximity more successfully.

国立情報学研究所で電子化したコンテンツを使用している。

<p>Several researchers have claimed that passively viewing manipulable objects results in automatic motor activation of affordances regardless of intention to act upon an object. Support for the <i>automatic activation account </i>stems primarily from findings using stimulus-response compatibility paradigms in which responses are fastest when there is correspondence between one’s response hand and an object’s handle. Counter to this view is the <i>spatial coding account</i>, which suggests that past findings are a result of abstract spatial codes stemming from salient object properties and their left-right correspondence with responses. Although there is now considerable support for this account, there has been little attention paid to determining whether evidence in favor of the automatic activation account will be evident after accounting for the spatial issues demonstrated by the spatial coding account.</p><p>The present study involved five experiments conducted to bridge this gap in two steps. First, I aimed to demonstrate the importance of considering spatial issues and left-right correspondence when studying object-based motor activation by numerous objects championed by past researchers who attempted to similarly address the aforementioned issue (Experiments 1 and 2). Second, I sought to determine whether evidence favoring the automatic activation account could be obtained when the possibility for left-right correspondence was absent in a novel set of stimuli created specifically for this purpose (Experiments 3, 4, and 5).</p><p>Experiment 1 examined a stimulus set that some researchers have suggested can more definitively tease apart evidence for automatic activation from the influence of spatial factors studies. Experiment 2 was more narrowly focused and investigated a single object presented in different horizontal orientations. These experiments effectively demonstrated the importance of giving more consideration to the nature of the stimuli used in object-based compatibility studies and how they are presented. The results of Experiment 1 suggest that a stimulus set that has been claimed to sidestep spatial confounds does not, in fact, do so. Moreover, Experiment 2 demonstrated that performance could be influenced by simple rotation of the object to which a response was required.</p><p>Having established the importance of controlling the stimuli used to investigate automatic activation of afforded responses, I turned to determining whether a novel stimulus set would yield findings favoring the automatic activation account even after accounting for left-right correspondence (Experiments 3, 4, and 5). Three sets of novel object stimuli were developed that do not allow for left-right correspondence and could iteratively assess support for the automatic activation account based on criteria for activation that have been put forth in the literature. The three sets of stimuli contained no information about shape nor functionality (i.e., silhouette iteration) or information about shape and functionality (i.e., functional iteration), or they were an intermediate between the two other types (i.e., intermediate iteration).</p><p>Critically, the three latter experiments progressively approached the conditions that researchers have suggested are ideal for automatic activation of afforded responses to occur. Experiment 3 tasked participants with completing a color discrimination task in which they viewed only one of the three object iterations and responded with button presses. Experiment 4 used the same experimental configuration, but instead, required participants to respond with a grasping response. Finally, Experiment 5 required participants to complete a reach-and-grasp response in an object discrimination task using both the silhouette and functional iterations.</p><p>Across Experiments 3, 4, and 5, no support for the automatic activation account of afforded responses was found. Although the automatic activation account would predict that individuals should be fastest at responding to the functional stimuli than to the other two object iterations, no such evidence was observed. Given that the possibility for left-right correspondence was removed from the novel stimulus set studied here, these results provide indirect support for the spatial coding account of prior results and further indicate that past findings favoring the automatic activation account have largely been a result of left-right correspondence. </p>

Abstract concepts seem to be related to space dimension. Evidence of this relation refers to the domain of numerical cognition. An example is the SNARC effect (Spatial Numerical Association of Response Codes, Dehaene, Bossini, and Giraux 1993), which consists in the observation that people react faster to small number with the left hand and to large number with the right hand. This number-space interaction has been explained according to the mental number line hypothesis (e.g. Restle 1970; Dehaene, Bossini, and Giraux 1993), which claims that the representation of numbers has the form of a horizontal line upon which numbers are represented from left to right. Recently, an alternative account suggests that the association between numbers and space results from a decision process to categorize numbers as “small” and “large” before being associated with space dimension (e.g. Gevers et al. 2006b, 2010; Van Opstal and Verguts 2013). The first goal of this thesis is investigating the spatial coding of numbers. In a first study, magnitude concepts such as “small” and “large” were observed to be spatially organized like numbers. In a second study, these magnitude concepts were intermixed with numbers in a reversal design (e.g. Notebaert et al. 2006). In this study, responding as incompatible to magnitude concepts with hand or foot was observed to reverse the spatial mapping of numbers, supporting the idea that the congruency between numbers and space results from conceptual coding of magnitude (e.g. Gevers et al. 2006b, 2010; see also Van Opstal and Verguts 2013). Further evidence of association between abstract concepts and space has been provided also in the domain of emotion. On one hand, Casasanto (2009a) demonstrated that people spontaneously associate positive valence with the side of space congruent to the dominant hand. On the other hand, Holmes and Lourenco (2011) observed that emotional expressions are left-to-right spatially organized with increasing in happiness/angriness rather than positive/negative valence. A second aim of this thesis is focused on investigating the spatial coding of emotion. This was meant to understand how general are the spatial mechanisms. In a third study, the reversal paradigm (e.g. Notebaert et al. 2006) was adopted to investigate the processing mechanism underlying spatial coding of numbers and emotional valence concepts. Manipulation of the mapping between valence concepts and lateralized responses did not influence the spatial coding of numbers, suggesting a separate underlying architecture. Finally, in a fourth study, spatial coding of emotion was observed according to both valence and arousal dimensions (Casasanto 2009a; Holmes and Lourenco 2011). / Doctorat en Sciences psychologiques et de l'éducation / info:eu-repo/semantics/nonPublished

Les saccades sont de brefs mouvements des yeux dont le but est d'amener les objets visuels périphériques sur la partie fovéale de la rétine. L'ensemble des modèles considère que la programmation de la métrique des saccades reflète tout d'abord le codage spatial distribué au sein du colliculus supérieur (CS), et n'est qu'ensuite modulée par des processus cognitifs endogènes. La majorité considère que les interactions latérales entre les neurones du CS (locales et excitatrices ou distantes et inhibitrices) déterminent où mais aussi quand les yeux bougent. Nos études visaient à (1) tester et préciser la relation entre codage spatial distribué et métrique des saccades, (2) re-examiner si des stratégies visuelles peuvent déterminer cette métrique, et (3) tester le rôle des interactions latérales. Elles reposaient sur l'enregistrement des mouvements oculaires de participants humains lors de la visée d'une cible visuelle, présentée seule ou accompagnée d'un distracteur. Nos résultats ont confirmé l'hypothèse de codage spatial distribué ; les champs d'intégration spatiale estimés à partir de la distance maximale entre deux stimuli pour l'exécution d'une saccade vers une position intermédiaire (ou effet global; Findlay, 1982) présentent des propriétés similaires aux champs récepteurs des neurones du CS. Deuxièmement, en désaccord avec l'hypothèse générale, des stratégies visuelles peuvent aussi amener le regard au centre de gravité. Enfin, contrairement à l'hypothèse d'interactions latérales, l'effet d'un distracteur sur la latence des saccades (Walker et al., 1997) est indépendant de la distance qui le sépare de la cible. / Saccades are brief movements of the eyes which bring peripheral visual objects onto the central, foveal part of the retina for detailed visual analysis. All models assume that the programming of saccade metrics primarily reflects distributed spatial coding in the Superior Colliculus (SC), and is only modulated by cognitive, endogenous processes. Furthermore, the majority of models rely on the assumption that lateral interactions between collicular neurons (local and excitatory or distant and inhibitory) are responsible for both where and when the eyes move. The present studies aimed at (1) testing and specifying the relationship between distributed spatial coding and saccade metrics, (2) re-examining the role of visual strategies on saccade metrics, and (3) testing the role of lateral interactions. To this aim, humans' eye movements were recorded in saccade-target tasks, in which the target was presented with or without a distractor stimulus. Results first confirmed the distributed spatial-coding hypothesis by showing that spatial-integration fields as estimated by the maximal distance between two stimuli for the eyes to move to an intermediate location (or global effect; Findlay, 1982) share the same properties as the receptive fields of collicular neurons. Second, in contradiction with the general assumption, visual strategies can also take the eyes to the centre of gravity of the global visual configuration. Third, contrary to the lateral-interaction hypothesis, the effect of a distractor on saccade latency (Walker et al., 1997) is independent of its distance to the target.

This thesis deals with machine learning techniques for the extraction of structure and the analysis of the vertebrate olfactory pathway based on related methods. Some of its main contributions are summarized below. We have performed a systematic investigation for classification in biomedical images with the goal of recognizing a material in these images by its texture. This investigation included (i) different measures for evaluating the importance of image descriptors (features), (ii) methods to select a feature set based on these evaluations, and (iii) classification algorithms. Image features were evaluated according to their estimated relevance for the classification task and their redundancy with other features. For this purpose, we proposed a framework for relevance and redundancy measures and, within this framework, we proposed two new measures. These were the value difference metric and the fit criterion. Both measures performed well in comparison with other previously used ones for evaluating features. We also proposed a Hopfield network as a method for feature selection, which in experiments gave one of the best results relative to other previously used approaches. We proposed a genetic algorithm for clustering and tested it on several realworld datasets. This genetic algorithm was novel in several ways, including (i) the use of intra-cluster distance as additional optimization criterion, (ii) an annealing procedure, and (iii) adaptation of mutation rates. As opposed to many conventional clustering algorithms, our optimization framework allowed us to use different cluster validation measures including those which do not rely on cluster centroids. We demonstrated the use of the clustering algorithm experimentally with several cluster validity measures as optimization criteria. We compared the performance of our clustering algorithm to that of the often-used fuzzy c-means algorithm on several standard machine learning datasets from the University of California/Urvine (UCI) and obtained good results. The organization of representations in the brain has been observed at several stages of processing to spatially decompose input from the environment into features that are somehow relevant from a behavioral or perceptual standpoint. For the perception of smells, the analysis of such an organization, however, is not as straightforward because of the missing metric. Some studies report spatial clusters for several combinations of physico-chemical properties in the olfactory bulb at the level of the glomeruli. We performed a systematic study of representations based on a dataset of activity-related images comprising more than 350 odorants and covering the whole spatial array of the first synaptic level in the olfactory system. We found clustered representations for several physico-chemical properties. We compared the relevance of these properties to activations and estimated the size of the coding zones. The results confirmed and extended previous studies on olfactory coding for physico-chemical properties. Particularly of interest was the spatial progression by carbon chain that we found. We discussed our estimates of relevance and coding size in the context of processing strategies. We think that the results obtained in this study could guide the search into olfactory coding primitives and the understanding of the stimulus space. In a second study on representations in the olfactory bulb, we grouped odorants together by perceptual categories, such as floral and fruity. By the application of the same statistical methods as in the previous study, we found clustered zones for these categories. Furthermore, we found that distances between spatial representations were related to perceptual differences in humans as reported in the literature. This was possibly the first time that such an analysis had been done. Apart from pointing towards a spatial decomposition by perceptual dimensions, results indicate that distance relationships between representations could be perceptually meaningful. In a third study, we modeled axon convergence from olfactory receptor neurons to the olfactory bulb. Sensory neurons were stimulated by a set of biologically-relevant odors, which were described by a set of physico-chemical properties that covaried with the neural and glomerular population activity in the olfactory bulb. Convergence was mediated by the covariance between olfactory neurons. In our model, we could replicate the formation of glomeruli and concentration coding as reported in the literature, and further, we found that the spatial relationships between representational zones resulting from our model correlated with reported perceptual differences between odor categories. This shows that natural statistics, including similarity of physico-chemical structure of odorants, can give rise to an ordered arrangement of representations at the olfactory bulb level where the distances between representations are perceptually relevant. / <p>QC 20120224</p>