La prise de décision est un mécanisme qui fait intervenir les structures neuronales supérieures afin d’effectuer un lien entre la perception du signal et l’action. Plusieurs travaux qui cherchent à comprendre les mécanismes de la prise de décision sont menés à divers ni- veaux allant de l’analyse comportementale cognitive jusqu'à la modélisation computationnelle. Le but de ce projet a été d’évaluer d’un instant à l’autre comment la variabilité du signal observé («bruit»), influence la capacité des sujets humains à détecter la direction du mouvement dans un stimulus visuel. Dans ces travaux, nous avons éliminé l’une des sources potentielles de variabilité, la variabilité d’une image à l’autre, dans le nombre de points qui portaient les trois signaux de mouvements cohérents (gauche, droite, et aléatoire) dans les stimuli de Kinématogramme de points aléatoires (KPA), c’est-à-dire la variabilité d’origine périphérique. Les stimuli KPA de type « V6 » étaient des stimuli KPA standard avec une variabilité instantanée du signal, et par contre les stimuli KPA de type « V8 », étaient modifiés pour éliminer la variabilité stochastique due à la variabilité du nombre de pixels d’un instant à l’autre qui portent le signal cohérent. Si la performance des sujets, qui correspond à leur temps de réaction et au nombre de bonnes réponses, diffère en réponse aux stimuli dont le nombre de points en mouvement cohérent varie (V6) ou ne varie pas (V8), ceci serait une preuve que la variabilité d’origine périphérique modulerait le processus décisionnel. Par contre, si la performance des sujets ne diffère pas entre ces deux types de stimuli, ceci serait une preuve que la source majeure de variabilité de performance est d’origine centrale. Dans nos résultats nous avons constaté que le temps de réaction et le nombre de bonnes réponses sont modulés par la preuve nette du mouvement cohérent. De plus on a pu établir qu’en éliminant la variabilité d’origine périphérique définit ci-dessus, on n’observe pas réellement de modification dans les enregistrements. Ce qui nous à amené à penser qu’il n y a pas de distinction claire entre la distribution des erreurs et les bonnes réponses effectuées pour chacun des essais entre les deux stimuli que nous avons utilisé : V6 et V8. C’est donc après avoir mesuré la « quantité d’énergie » que nous avons proposé que la variabilité observée dans les résultats serait probablement d’origine centrale. / Decision-making is a mechanism that primarily involves supraspinal neural structures to perform a link between the perception of the signal and action. Several studies that seek to understand the mechanisms of decision-making are conducted at various levels ranging from cognitive behavioral analysis to computational modeling. The purpose of this project was to evaluate how moment-to-moment variability of the sensory stimuli observed, influences the ability of humans to detect the direction of motion in a visual stimulus. In this work, we eliminated one of the potential sources of variability, namely variability from one visual image to another of the number of points which carried three different motion signals (coherent left, coherent right, and random) in Random-dot Kinematogram (RDK) stimuli, i.e a source of variability of external origin. “V6” stimuli were standard RDK stimuli with instantaneous (i.e., moment-to-moment) signal variability, whereas “V8” stimuli were modified to eliminate the stochastic variability due to variability in the number of points, which carried the coherent signal from image frame to image frame in the RDK stimuli. If the performance of the subjects, as measured by their reaction times and the number of correct answers, differs in response to the stimuli of which the number of points moving coherently varies (V6) or does not vary (V8) from moment to moment in the visual stimulus, this would be a proof that the variability of peripheral origin would modulate the decision-making process. On the other hand, if the performance of the subjects does not differ between the two types of stimuli, this would be a proof that the major source of variability of performance is of central origin. In our results we found that the strength of coherent movement modulates the reaction times and the number of correct responses of subjects. However, eliminating the variability of peripheral origin defined above, had little significant effect on the performance of the subjects. There is no clear difference between the distribution of reaction times or between errors and correct answers for each test performed between both V6 and V8 stimuli. Finally, after measuring the amount of motion energy in the RDK stimuli, we could propose that: the origin of the observed variability in the results would primarily be of central origin.

GONÇALVES, Fabrício Mota. Evapotranspiração e coeficientes de cultivo (KC) da cana-de-açúcar irrigada por gotejamento subsuperficial. 2010. 64 f. : Dissertação (mestrado) - Universidade Federal do Ceará, Centro de Ciências Agrárias, Departamento de Engenharia Agrícola, Programa de Pós-Graduação em Engenharia Agrícola, Fortaleza-CE, 2010. / Submitted by demia Maia (demiamlm@gmail.com) on 2016-06-28T16:59:16Z No. of bitstreams: 1 2010_dis_fmgonçalves.pdf: 3900094 bytes, checksum: 1ad79acf1432c0de09bcbdf52cb0f8c3 (MD5) / Approved for entry into archive by demia Maia (demiamlm@gmail.com) on 2016-06-28T16:59:48Z (GMT) No. of bitstreams: 1 2010_dis_fmgonçalves.pdf: 3900094 bytes, checksum: 1ad79acf1432c0de09bcbdf52cb0f8c3 (MD5) / Made available in DSpace on 2016-06-28T16:59:48Z (GMT). No. of bitstreams: 1 2010_dis_fmgonçalves.pdf: 3900094 bytes, checksum: 1ad79acf1432c0de09bcbdf52cb0f8c3 (MD5) Previous issue date: 2010 / Sugar cane is one of the crops with the highest level of water consumption, with a high evapotranspiration demand throughout most of its cycle. Knowledge of the sugar cane water demand, the time that irrigation should be applied, and how to irrigate the crop are becoming increasingly important for the sustainable development of irrigated sugar cane cultivation, especially in areas of low water availability, as it is the case in northea stern Brazil. The experiment was carried out at the Curu Experimental Field (property of the Embrapa Tropical Agroindustry), located in the Curu-Paraipaba Irrig ation Project, city of Paraipaba-CE (3 ° 29 '20''S, 39 ° 9' 45''W and elevation 30m), in order to determine the evapotrans piration and crop coefficients (Kc) of sugarcane (Saccharum L. officinarium), subsurface drip-irrigated. It was used the SP 6949 sugar cane variety, planted in a spacing of 1.8 m between double rows and 0.4 m between rows within the double row. As to t he irrigation, we used one lateral water line for each double row, buried at a depth of 0.15 m, w ith self-compensating drip emitters, with a flow rate of 1 L h-1, spaced 0.5 m on the water line. The crop was irrigated daily and there were three fertigations per week. The irrigation de pth was periodically adjusted, keeping the soil water potential between -8 kPa to -20 kPa. The crop evapotranspiration (ETc) was determined using a wei ghing lysimeter with a surface area of 2.25 m 2. The reference evapotranspiration (ETo) was estimated by the FAO Penman-Monteith method. The durations of the phenological stages of cane sugar were estimated through analysis of soil cover, with the use of digital images. Durations of 31, 49, 237 and 118 days were observed, respectively for initial, crop development, mid-season and late season stages. The yields achieved within and outside the lysimeter were respectively 144.4 and 108.8 t ha -1 . The total ETc observed during the cycle of the sugar cane was 1074.1 m m, with maximum values of 6.6 mm d-1 during the mid-season stage. Observed Kc values for initial, mid-season and late season stages were respectively 0.23, 1 .03 and 0.50. In the crop development stage, the relationship between Kc and the (crop) soil cover was represented with good accuracy by a negative quadratic model. / A cana-de-açúcar é uma das culturas com maior consumo de água, apresentando uma alta demanda evapotranspirativa ao longo da maior parte do seu ciclo. O conhecimento das necessidades hídricas da cana-de-açúcar, do momento de aplicação da água requerida pela mesma e a forma de aplicação torna-se cada vez mais importante para o desenvolvimento sustentável da sua produção irrigada, principalmente em regiões de pouca disponibilidade hídrica, como é o caso da região Nordeste do Brasil. O experimento foi conduzido no Campo Experimental do Curu, pertencente à Embrapa Agroindústria Tropical localizado no Perímetro Irrigado Curu-Paraipaba, município de Paraipaba-CE (3° 29’ 20’’ S, 39° 9’ 45’’ W e altitude de 30 m), com o objetivo de determinar a evapotranspiração e os coeficientes de cultivo (Kc) da cana-de-açúcar (Saccharum officinarium L.), irrigada por gotejamento subsuperficial. Utilizou-se a variedade de cana SP 6949, plantada no espaçamento de 1,8 m entre fileiras duplas e 0,4 m entre linhas dentro da fileira dupla. Na irrigação utilizou-se uma linha lateral para cada fileira dupla, enterrada na profundidade de 0,15 m, com gotejadores autocompensantes, com vazão de 1 L h-1, espaçados de 0,5 m na linha. A cultura foi irrigada com freqüência diária e foram realizadas três fertirrigações por semana. A lâmina de irrigação foi ajustada periodicamente, mantendo-se o potencial de água no solo entre -8 kPa e -20 kPa. A evapotranspiração da cultura (ETc) foi determinada utilizando-se um lisímetro de pesagem com área superficial de 2,25 m2. A evapotranspiração de referência (ETo) foi estimada pelo método FAO Penman-Monteith. As durações das fases fenológicas da cana-de-açúcar foram estimadas por meio da análise da cobertura do solo pela cultura, determinada com o uso de imagens digitais. Foram observadas durações de 31, 49, 237 e 118 dias, para as fases inicial, de desenvolvimento, intermediária e final, respectivamente. A produtividade alcançada dentro e fora do lisímetro foi de 144,4 e 108,8 t ha-1, respectivamente. A ETc total observada durante o ciclo da cana-de-açúcar foi de 1.074,1 mm, com valores máximos da ordem de 6,6 mm d-1 durante a fase intermediária. Foram observados valores de Kc iguais a 0,23, 1,03 e 0,50, para as fases inicial, intermediária e final, respectivamente. Na fase de desenvolvimento vegetativo a relação entre o Kc e a cobertura do solo pela cultura foi representada com bastante exatidão por um modelo quadrático negativo.