No description available.

The Stroop effect is commonly demonstrated by measuring the time required to identify the colour of the ink (e.g. blue) in which a colour word (e.g. RED) is printed and illustrates a strong interference effect when the prepotent response of processing the word must be inhibited to respond correctly to the colour of the ink. Longer response latencies in colour-identification tasks involving emotional words versus neutral words, an effect dubbed the Emotional Stroop (ES), is often likened to the type of interference found in the Stroop task. However, research has suggested that the effect of attentional modulation in Stroop tasks may be very different from the kind of emotional interference found when comparing reaction times to emotional versus neutral words (McKenna & Sharma, 2004). Proportional congruency experiments using the Stroop task manipulate the amount of incongruity present in a block of trials (Logan & Zbrodoff, 1979), and demonstrate attentional modulation as a change in the size of the Stroop effect such that a greater amount of Stroop interference in a block is associated with a smaller Stroop effect. Manipulating the proportion of interference trials may inform understanding of the differences between Stroop colour-word interference and emotional-word interference. In three experiments, we manipulated the proportion of emotional words in a mixed-list design to study the effect of proportion manipulation on the ES. An enlarged ES was found in blocks of trials that contained more emotional interference; a finding contrary to attentional modulation seen for proportion manipulations of congruency in the Stroop task. The differences between the ES effect and the Stroop effect are discussed, including the role of response incongruity as one possible reason for the discrepancy. / Thesis / Master of Science (MSc) / In three experiments, we investigated the role of proportionally more or less emotional interference on a colour naming task. Emotional interference in the form of emotional words was compared to response interference where participants see mismatched and matched colour words (e.g. the word ‘NIGHTMARE’ in green versus the word ‘BLUE’ in green). We expected participants to be able to learn which blocks contained more emotional word content and to adapt in order to reduce this slow-down effect. Instead, participants were especially slow on emotional words when they were in blocks that contained more emotional word interference. The possible importance of emotional stimuli to participants over the task demands of naming the colour of a word, as compared to other kinds of interference, is discussed.

Reabilitação robótica tem um papel importante em exercícios terapêuticos ao combinar robôs com jogos sérios de computador em uma atraente plataforma terapêutica. Entretanto, a tarefa de medir o grau de adesão do paciente ao tratamento não é trivial. A dificuldade de aplicar técnicas baseadas em questionários e entrevistas, particularmente em pacientes que tiveram a fala comprometida por acidente vascular encefálico (AVE), nos inspirou a investigar técnicas não verbais e não invasivas para classificar emoções. Com este propósito, uma rede neural supervisionada foi projetada para interpretar imagens térmicas infravermelhas faciais de indivíduos realizando terapia robótica de reabilitação integrada com os jogos. Uma base de dados contendo imagens de 8 voluntários foi criada e contém reações emocionais espontâneas e provocadas. No total, foram analizadas 2445 imagens térmicas faciais, em média 100 imagens por pessoa por 3 categorias de emoções (neutra, motivado e sobrecarregado). Baseado em análise de matriz de confusão, os resultados experimentais se correlacionaram muito bem com as estimativas manuais, produzindo um desempenho global de 92,6%. / Rehabilitation robotic plays an important role in therapeutic exercises by combining robots with computer serious games into an attractive therapeutic platform. However, measuring the degree of engagement of the user is not a trivial task. The difficulty of applying question-based techniques, particularly for patients who have the speech capacity compromised due to cerebrovascular accidents, has inspired us to investigate noninvasive and nonverbal techniques aiming to classifying emotions. For this purpose, a supervised artificial neural network interprets facial infrared thermal images of individuals performing rehabilitation robotic therapy integrated with games. A database containing images of 8 users was generated by combining evoked and spontaneous emotional reactions. In total, 2445 facial thermal images with an average of 100 images per person for three categories of emotions (neutral, motivated, and overstressed) were analyzed. Based on confusion matrix analysis, the experimental results correlated very well with manual estimates, producing an overall performance of 92.6%.

Reabilitação robótica tem um papel importante em exercícios terapêuticos ao combinar robôs com jogos sérios de computador em uma atraente plataforma terapêutica. Entretanto, a tarefa de medir o grau de adesão do paciente ao tratamento não é trivial. A dificuldade de aplicar técnicas baseadas em questionários e entrevistas, particularmente em pacientes que tiveram a fala comprometida por acidente vascular encefálico (AVE), nos inspirou a investigar técnicas não verbais e não invasivas para classificar emoções. Com este propósito, uma rede neural supervisionada foi projetada para interpretar imagens térmicas infravermelhas faciais de indivíduos realizando terapia robótica de reabilitação integrada com os jogos. Uma base de dados contendo imagens de 8 voluntários foi criada e contém reações emocionais espontâneas e provocadas. No total, foram analizadas 2445 imagens térmicas faciais, em média 100 imagens por pessoa por 3 categorias de emoções (neutra, motivado e sobrecarregado). Baseado em análise de matriz de confusão, os resultados experimentais se correlacionaram muito bem com as estimativas manuais, produzindo um desempenho global de 92,6%. / Rehabilitation robotic plays an important role in therapeutic exercises by combining robots with computer serious games into an attractive therapeutic platform. However, measuring the degree of engagement of the user is not a trivial task. The difficulty of applying question-based techniques, particularly for patients who have the speech capacity compromised due to cerebrovascular accidents, has inspired us to investigate noninvasive and nonverbal techniques aiming to classifying emotions. For this purpose, a supervised artificial neural network interprets facial infrared thermal images of individuals performing rehabilitation robotic therapy integrated with games. A database containing images of 8 users was generated by combining evoked and spontaneous emotional reactions. In total, 2445 facial thermal images with an average of 100 images per person for three categories of emotions (neutral, motivated, and overstressed) were analyzed. Based on confusion matrix analysis, the experimental results correlated very well with manual estimates, producing an overall performance of 92.6%.

Die zentrale Frage der Arbeit „Wie beeinflussen Emotionen kognitive Prozesse?“ wurde in den letzten Jahrzehnten intensiv erforscht. Dabei stellte sich heraus, dass entgegen der alltagspsychologischen Annahme, Emotionen nicht ausschließlich negative Effekte auf die kognitiven Leistungen haben. So fand zum Beispiel Bower (1981) sehr spezifische Einflüsse von Stimmungen auf das Gedächtnis und Isen (1999) berichtete Verbesserungen der kognitiven Leistungsfähigkeit unter positivem Gefühlseinfluss in unterschiedlichen Aufgaben. Fasst man die aktuelle Befundlage insgesamt zusammen, so ist es jedoch unmöglich ein eindeutiges Fazit zu ziehen, so heterogen sind die Ergebnisse der unterschiedlichen Studien (Rusting, 1998). Zunächst werden in dieser Arbeit daher zwei mögliche Ursachen für das heterogene Befundmuster diskutiert. Zum einen führt die Dominanz des dimensionalen Ansatzes (Russell, 2003; Bradley und Lang, 2000; Davidson et al., 2003) in der Emotionspsychologie dazu, dass spezifische Basisemotionen mit derselben Valenz und Erregung nicht unterschieden werden (Ekman, 1999; Izard, 1977; Plutchik, 1962). Zum anderen werden Emotion und Kognition meist als separate Prozesse betrachtet (Lazarus, 1984; Zajonc, 1980). In letzter Zeit mehren sich jedoch die Stimmen, die sich für eine integrative Perspektive auf die Interaktion von Emotion und Kognition aussprechen (Gray, 2004; Lewis, 2005; Scherer, 2000). Daher wird in dieser Arbeit vorgeschlagen, Emotionen nicht als separaten Elemente der menschlichen Psyche zu betrachten, sondern als spezifische Muster, die sich aus verschiedenen Komponenten zusammen setzen: u.a. Physiologie, Kognition, Motivation. Entsprechen einige dieser Komponenten einem bestimmten Emotionsmuster, so können in der Folge die übrigen Komponenten durch das Muster selbst angeregt werden und zur vollen Entfaltung des Emotionsmusters führen. Um dieses Modell zu testen, sollte in dieser Arbeit die Hypothese geprüft werden, ob für spezifische Emotionen spezifische Muster der Ausprägung kognitiver Parameter existieren. Die empirischen Untersuchungen dieser Arbeit konzentrierten sich dabei auf drei spezifische Basisemotionen (Angst, Ärger, Traurigkeit) und auf zwei ausgewählte kognitive Parameter, die an Arbeiten von Dörner (1999, Dörner und Bartl-Storck, 2002) angelehnt sind: Wechselbereitschaft (Flexibilität) und Assoziationsbreite (Kreativität). Die Stimmungsinduktion erfolgte neben dem Lesen von Geschichten vorwiegend mittels der Methode der Imagination eigener Erlebnisse. Um die beiden kognitiven Parameter Wechselbereitschaft und Assoziationsbreite zu messen, wurden zwei Paradigmen eingesetzt. Bei dem ersten handelte es sich um ein Aufgaben-Wechsel-Paradigma nach Dreisbach und Goschke (2004). Ärgerliche Probanden zeigten darin Schwierigkeiten beim Aufgabenwechsel, was auf eine verringerte Wechselbereitschaft schließen lässt. Das zweite Paradigma war ein Assoziations-Erkennungs-Paradigma, das in dieser Arbeit neu entwickelt wurde, um beide Parameter innerhalb ein und desselben Paradigmas zu messen. Es fand sich darin konsistent eine Erhöhung der Assoziationsbreite in ängstlicher Stimmung. Die Stimmungen zeigten im Gegensatz zum ersten Paradigma keine Unterschiede in ihrer Auswirkung auf die Wechselbereitschaft. Zusammenfassend geht aus der Arbeit hervor, dass es spezifische Effekte spezifischer Stimmungen gibt, es also nicht genügt, Valenz und Arousal zu unterscheiden. Allerdings sind auch die in dieser Arbeit gefundenen Effekte nicht immer stabil. Dies wird zum Teil auf zu schwache Stimmungsveränderungen, zu geringe Stichprobengröße und Stimmungsregulationsprozesse zurückgeführt. Bezogen auf die allgemeine Frage nach dem Einfluss von Emotionen auf kognitive Prozesse präsentiert diese Arbeit einen neuen Forschungsansatz, der sowohl eine theoretische Präzisierung der abstrakten Konzepte als auch eine methodische Integration der vielfältigen Befunde erlaubt. Dies geschieht mittels der Idee von einer Emotion als einen das gesamte System Mensch erfassenden Prozess. Damit legt diese Arbeit eine Basis, auf der sich zukünftige Forschungsarbeiten weiter der Frage annähern können, ob und wie unsere Gefühle zu unseren oft so beeindruckenden Verstandesleistungen beitragen.

The ability to prioritize information enables us to think and take action without being overwhelmed by external stimuli or internal thoughts and feelings. Neuroticism is associated with altered processing of emotional information but differences in the processing of emotional information may arise from basic differences in information processing, such as altered processes of attention, changes in sensitivity to salient information, or differences in the ability to encode conjunctions of information. Through this thesis, I explore the relationship between neuroticism and processing of non-emotional information, with a particular focus on learning about combinations of information. Associative learning paradigms were used to test ability to learn about combinations of information and neuroticism was observed to be associated with strong non-linear discrimination learning. The tendency to focus on specific details was associated with weak non-linear discrimination learning. A novel model of associative learning is presented, offering an account for how variation in the ability to engage in non-linear discrimination learning might be understood. Mechanisms underlying the association between neuroticism and strong non-linear discrimination learning were explored. Neuroticism was not found to be associated with a tendency to focus on specific details or shifts in attention towards goal relevant information. Neuroticism was not found to be associated with enhanced ability to identify feature conjunctions, altered sensitivity to the relative validity of stimuli or pre-exposure of stimuli. The importance of understanding individual differences in processes of associative and the value of associative learning tasks to look at information processing biases underlying neuroticism are discussed.