Att spela ett stycke utan noter är en viktig aspekt i musicerandet, men trots detta finns få konkreta riktlinjer för hur memoreringsprocessen optimeras, och musiker hittar ofta sin egen väg. Syftet med denna studie var därmed att, ur ett designteoretiskt perspektiv, undersöka vilka resurser som ligger till grund för effektiv memorering. För att undersöka detta testades två memoreringsmetoder, konstruerade i linje med tidigare litteratur och forskning, på två Intermezzo av Brahms. Metod ett baserades på mekanisk inlärning utan analys, och metod två baserades på harmonikanalys. Videoobservation och loggbok användes för att dokumentera arbetet för varje metod, och effektiviteten mättes genom antal använda timmar. Resultatet visade att metod två i större utsträckning utnyttjade konceptuellt ramverk jämfört med metod ett, men att metoderna i övrigt liknade varandra i använda resurser. Metod två (5 timmar 15 minuter) var dessutom mer tidseffektiv jämfört med metod ett (14 timmar 20 minuter). I tillägg till den förbättrade effektiviteten bidrog dessutom metod två till större medvetenhet i musicerandet, vilket i sin tur ledde till ökad motivation för memorering. I diskussionen undersöks studiens resultat i relation till tidigare litteratur och forskning. / Playing a piece by heart is an important aspect of being a musician, despite this there are few concrete guidelines for how the memorization process is optimized, and musicians often find their own way. The purpose of this study was therefore to, from a design theory perspective, investigate which resources underlie effective memorization. To investigate this, two memorization methods, constructed in line with previous literature and research, were tested on two Intermezzo by Brahms. Method one was based on mechanical learning without analysis, method two was based on harmonics analysis. Video observation and logbook were used to document the process, efficiency was measured by utilized hours. Results showed that method two depended more on conceptual framework compared to method one, the methods were however similar in other used resources. Method two (5 hours 15 minutes) was more time-efficient than method one (14 hours 20 minutes). In addition to the improved efficiency, method two contributed to greater awareness in the memorization process, which in turn led to increased motivation for memorization. The discussion examines the study's results in relation to previous literature and research.

En studie om medveten memorering av harmonik inom klassiskt pianospel. Arbetet bygger på observationer av en tolv veckors lång instuderingsprocess, sett i relation till ett urval av tidigare litteratur och forskning inom området. Studien har i syfte att ge en djupare bild och förståelse av memoreringens påverkan vid konsertsituationen samt vilka fördelar respektive nackdelar det kan finnas med medveten memorering av harmonik. För att uppnå syftet tar studien utgångspunkt i tre frågeställningar. Frågeställningarna berör medveten memorering av harmonik som metod att instudera ett stycke, vilka fördelar och nackdelar det kan få samt om det kan hjälpa en musicerande att känna sig tryggare på scen. Som blivande lärare hoppas jag att detta dels ska hjälpa mig vid egna uppspel men också att jag ska kunna förmedla en bra memoreringsmetod till de elever som jag kommer att möta under min tid som lärare. / A study about conscious memorization of harmony in classical piano playing. The study is based on observations of a twelve week period and on chosen literature from previous research in this field. The purpose of the study is to get a deeper knowledge and understanding in how memorizing can effect the results at a performance and also what other benefits and disadvantages there could be. The study has as a starting point three questions. These questions are about memorization of harmonics as a method to learn a music piece, the benefits and disadvantages with conscious memorization and also if a pianist can feel safer on stage after consciously memorizing harmony. As a future teacher I hope this study will help me with my own performances and also the pupils that I will teach and meet during my professional work as a musicteacher.

L’écran est une interface omniprésente utilisée quotidiennement. Sur-sollicité par les écrans, l’individu ne peut pas percevoir l’intégralité des informations. La compréhension de l’attention portée aux contenus médiatisés est un enjeu crucial dans la stratégie marketing. Un paramètre n’a pas été pris en compte dans les précédentes études: la qualité technique de l’image médiatisée. Elle peut être définie comme l’ensemble des paramètres suivants : la définition, la fréquence image, la luminosité et la couleur. Pour mesurer la perception de l’amélioration de ces paramètres, une phase exploratoire a été menée comprenant un test de qualité perçue, un focus group et des entretiens. Une phase empirique confirmatoire a suivi pour étudier l’attention visuelle et la mémorisation de l’information médiatisée grâce à deux expérimentations. Une série d’entretien et un focus group ont permis d’étudier l’hyperréalité dans l’image médiatisée et son impact sur l’immersion et la qualité d’expérience. / The screen is a ubiquitous interface daily used. Overextended by screens, the individual cannot perceive all the information. Understanding the visual attention to media content is a crucial issue in the marketing strategy. A parameter of the mediated image has not been taken into account in the studies conducted so far: the technical quality of the image. It can be defined as the set of parameters that objectively compose an image: the definition, the image frequency, the brightness dynamics and the color. To measure the perception of the improvement of these parameters, an exploratory phase was conducted including a perceived quality test, a focus group and interviews. A confirmatory empirical phase followed to study the visual attention and memorization of mediated information through two experiments. A series of interviews and a focus group explored hyperreality in the mediated image and its impact on immersion and quality of experience.

This study was carried out in order to discover new strategies to memorize piano music. There are six different types of memory involved in performing: auditory, kinesthetic, visual, analytical, nominal and emotional. There are two main ways of practicing: playing practice and non-playing practice. I tried to find out if the order in which we use these two kinds of practice affects the quality of the memorization. During one week I practiced three different pieces following three different methods: 1 Using only playing practice; 2 using first playing practice and then non-playing practice; 3 using first non-playing practice and then playing practice. The second method had a much better result than the other two. The whole process was registered with a video camera and a logbook. / <p>The exam concert is archived for copyright reasons until 2023.</p>

It is an important question in machine learning to understand how neural networks learn. This thesis sheds further light onto this by studying autoencoder neural networks which can memorize data by storing it as attractors.What this means is that an autoencoder can learn a training set and later produce parts or all of this training set even when using other inputs not belonging to this set. We seek out to illuminate the effect on how ReLU networks handle memorization when trained with different setups: with and without bias, for different widths and depths, and using two different types of training images -- from the CIFAR10 dataset and randomly generated. For this, we created controlled experiments in which we train autoencoders and compute the eigenvalues of their Jacobian matrices to discern the number of data points stored as attractors.We also manually verify and analyze these results for patterns and behavior. With this thesis we broaden the understanding of ReLU autoencoders: We find that the structure of the data has an impact on the number of attractors. For instance, we produced autoencoders where every training image became an attractor when we trained with random pictures but not with CIFAR10. Changes to depth and width on these two types of data also show different behaviour.Moreover, we observe that loss has less of an impact than expected on attractors of trained autoencoders.

Detta är ett självständigt arbete som handlar om hur jag utforskar min egen övning där jag memorerar texter ur två olika sånger under två separata veckor. Arbetet genomförs ur ett sociokulturellt perspektiv och grundar sig i tidigare forskning och självständiga arbeten om minne och musicerande utantill. Arbetet har dokumenterats genom videoobservationer och loggboksanteckningar under övningstillfällen hösten 2022. Resultatet presenterar olika teman som observerats och svarar där igenom på mina frågeställningar vilka redskap jag använder mig av och vilka faktorer som påverkar svårighetsgraden på lärandet av olika texter. I diskussionen sätter jag resultatet i relation till tidigare litteratur kring ämnet, forskning och det sociokulturella perspektivet. / This is a self-study about how I study my own practice where I memorized two different song lyrics in two separate weeks. The study is made with a sociocultural perspective and is based on previous research and studies about memory and playing music without sheet music. The study was being documented through video recordings and notes in a logbook in the autumn of 2022. The result presents different themes that came to be observed and they answer my research questions about which cultural tools I use and what it is that affects the degree of difficulty when learning song lyrics. In the discussion I put my result in relation to the previous literature, research and the sociocultural perspective.

Syftet med denna studie är att förbättra min förmåga att sjunga utantill för att därigenom utveckla min konstnärliga uttrycksförmåga. För att göra detta utvecklades en egen metod för utantillinlärning av sångtext genom att två långa visor av Evert Taube studerades in: Möte i monsunen och Eldareval- sen. En litteratur- och intervjustudie användes som grund för val av inlärningsmetoder. Experimentet utfördes i två delar. I första delen av experimentet undersöktes fyra metoder för utantillinlärning av text; (1) Nedskrivning av texten för hand, (2) Skandering av texten (rytmisk läsning), (3) Gestaltning av texten med rörelser och slutligen (4) en kombination av de tre första. I andra delen av experimen- tet användes erfarenheterna av första delen för att förfina metoden. I experimentets båda delar gjor- des även enkla textanalyser som komplement till övriga metoder.  Experimentet lades upp så att jag under fem veckor övade ovan nämnda visor med ovan nämnda me- toder under två dagliga pass på 25 minuter. Erfarenheterna av detta dokumenterades med 22 logg- boksinlägg och 19 videoinspelningar.  Resultaten tydde på att den bästa metoden för mig var en kombination av de fyra metoderna där skandering av texten var den viktigaste i kombination med analys av texten. Beroende på textens in- nehåll kunde sedan nedskrivning av text och gestaltning med rörelser användas i olika grad. Resulta- ten verifierade också rådande forskning som säger att enbart studier av minnets struktur och upp- byggnad i sig själv är användbart för att förbättra förmågan till memorering av musik. / The aim of this study is to increase my capacity to sing by heart in order to increase my ability of ar- tistic expression. To do this, a method for text memorization was developed through the learning of two long songs by Evert Taube: Möte i monsunen and Eldarevalsen. Firstly, an interview- and litera- ture study was done. This was the foundation for the selection of methods that were used in the ex- periment. Four methods were applied in the first part of the experiment: (1) Handwriting of the text, (2) Rhythmical chanting of the text, (3) embodying of the text using movements and lastly (4) a com- bination of the previous three. In the second part of the experiment, the experience from the first part was applied in order to refine the method. In both parts of the experiment, simple text analyses were performed as a complement to the other methods.  The experiment was done during five weeks where I practiced the above-mentioned songs and meth- ods daily for two sessions of 25 minutes. The experiences from this were documented through jour- naling and video documentation.  The results indicated that the best method for me was a combination of all four methods where rhyth- mical chanting should be the main method in combination with a proper analysis of the text. Depend- ing on the content of the text the other two methods could be applied. The results also verified the current research where study of the memory on its own could increase the capacity of music memori- zation.

Moderna barn använder mobilteknik oftare och oftare. Ett antal studier hävdar att barns lärande kan dra nytta av användning av medier i klassrum och att det finns en brist på denna teknik att använda i skolorna. Många undersökningar indikerar att användningen av augmented reality (AR) teknik i undervisning för barn hjälper dem att bli mer motiverade och engagerade i lärande activiteter, och att det är fördelaktigt för barns minneslagring. Vissa studier hävdar att motivation är en viktig faktor för att främja en inlärningsförmåga.Denna studie syftar på att skapa en augmented reality-bok för barn 4-5 år och på att förstå hur AR kan påverka engagemang, och hur den kan stödja lärandet i termer av memorering.Ett experiment genomfördes med åtta barn som går till förskolan. Under experimentet gjordes observationer och intervjuer. Grafisk design, illustration och animering var en del av förberedelserna för experimentet. Studieresultaten analyseras och diskuteras i förhållande till tidigare undersökningar i detta område.Resultaten pekar mot att AR skulle kunna hjälpa barn i förskoleåldern i klassrummsundervisning genom att positivt påverka deras minneslagring samt öka deras engagemang och motivation - som tidigare påstås är en mycket viktig faktor för att förbättra inlärningsförmåga. Samtidigt är det svårt att tolka resultaten med en hög grad av säkerhet för att testgrupperna var små, och antalet ytterligare faktorer kan ha påverkat resultaten. För att bekräfta resultaten av denna studie är det rekommenderat att genomföra en undersökning av en större skala. / Modern children use mobile technologies more and more often. A number of studies argue that children’s learning can benefit from the use of media in classrooms and that there exists a lack of these technologies applied at schools. Many research indicate that use of augmented reality technologies in children’s teaching helps children to get more motivated and engaged in the learning activities, and it is beneficial for children’s memory retention. Some studies claim that motivation is an essential factor for promotion a learning performance. This study aims at creating an augmented reality (AR) book for 4-5 years old children and on understanding how AR can affect engagement, and how it can support learning in terms of memorization.There was conducted an experiment with eight children who go to preschool. During the experiment, the observations and interviews were made. Graphic design, illustration and animation were a part of preparation for the experiment. The study results are analyzed and discussed in relation to earlier research in this area. The results point towards the fact that AR could help children of preschool age in classroom education by positively affecting their memory retention and increasing their engagement and motivation – which was earlier claimed to be an extremely important factor for improving the learning performance. At the same time, it is hard to interpret the results with a high degree of certainty as the test groups were small and a number of additional factors could have affected the results. In order to confirm the results of this study, it is suggested to conduct a study of a larger scale.

As interactive systems are now used to perform a variety of complex tasks, users need systems that are at the same time expressive, efficient and usable. Although simple interactive systems can be easily usable, interaction designers often consider that only expert practitioners can benefit from the expressiveness of more complex systems. Our approach, inspired by studies in phenomenology and psychology, underscores that non-experts have sizeable knowledge and advanced skills related to various expert activities having a social dimension -such as artistic activities-, which they gain implicitly through their engagement as perceivers. For example, we identify various music-related skills mastered by non-musicians, which they gain when listening to music or attending performances. We have two main arguments. First, interaction designers can reuse such implicit knowledge and skills to design interaction techniques that are both expressive and usable by novice users. Second, as expert artifacts and expert learning methods have evolved over time and have shown efficient to overcome the complexity of expert activities, they can be used as a source of inspiration to make expressive systems more easily usable by novice users. We provide a design framework for studying the usability and expressiveness of interaction techniques as two new aspects of the user experience, and explore this framework with three projects. In the first project we study the use of rhythmic patterns as an input method, and show that novice users are able to reproduce and memorize large vocabularies of patterns. This is made possible by the natural abilities of non-musicians to perceive, reproduce and make sense of rhythmic structures. We define a method to create expressive vocabularies of patterns, and show that novice users are able to efficiently use them as command triggers. In the second project, we study the design and learning of chording gestures on multitouch screens. We introduce design guidelines to create expressive chord vocabularies taking the mechanical constraints and the degrees of freedom of the human hand into account. We evaluate the usability of such gestures in an experiment and we present an adapted learning method inspired by the teaching of chords in music. We show that novice users are able to reproduce and memorize our vocabularies of chording gestures, while our learning method can improve long-term memorization. The final project focuses on music software used for live performances and proposes a framework for designing "instrumental" software allowing expert musical playing and having its elementary functionalities accessible to novices, as it is the case with acoustic instruments (for example, one can easily play a few chords on a piano without practice). We define a design framework inspired by a functional decomposition of acoustic instruments and present an adapted software architecture, both aiming to ease the design of such software and to make it match with instrument-making. These projects show that, in these cases: (i) the implicit knowledge novices have about some expert activities can be reused for interaction; (ii) expert learning methods can inspire ways to make expressive systems more usable novices; (iii) taking expert artifacts as a source of inspiration can help creating usable and expressive interactive systems. In this dissertation, we propose the study of usability as an alternative to the focus on immediacy that characterizes current commercial interactive systems. We also propose methods to benefit from the richness of expert activities and from the implicit knowledge of non-experts to design interactive systems that are at the same time expressive and usable by novice users.

Le 21ème siècle est considéré comme l’âge du numérique. Les TIC sont devenus des outils identitaires d’une génération désormais définie comme « native du numérique » (angl. digital natives). Equiper les établissements secondaires en matériels coûteux devient une nécessité pour faire face aux besoins nouveaux de la société. Les ressources numériques, qui ont complété puis évincé le manuel et ses périphériques traditionnels, demeurent un atout incontournable pour l’apprentissage d’une langue vivante étrangère, grâce à des contenus linguistiques et culturels le plus souvent authentiques, qui favorisent la mise en œuvre de l’approche actionnelle du Cadre Européen Commun de Référence pour les Langues. On peut néanmoins s’interroger sur la capacité du sujet apprenant à traiter cet input directement extrait d’Internet. La plus-value pédagogique présumée, didactique et motivationnelle, qui associée à ces outils technologiques ne saurait être acceptée sans un questionnement sincère sur les capacités cognitives effectives des apprenants ainsi que sur la réalité de leur engagement. L’accordage entre le sujet apprenant, les méthodes et méthodologies d’enseignement-apprentissage des langues, ainsi que les dispositifs numériques doivent être explorés. Il est nécessaire aujourd’hui de se situer dans un paradigme cognitiviste ou encore connexioniste en envisageant une construction de l’information et l’exposition des apprenants à un environnement stimulant et riche. En effet, en dépit de l’attractivité, de la non-linéarité et de l’interactivité de l’hypermédia, il convient de vérifier que l’apprenant possède bien les ressources cognitives nécessaires pour traiter efficacement l’information issue de l’hypertexte. Car l’enjeu du traitement de l’information ne se situe pas simplement au niveau de l’extraction d’input mais bien de la construction sémantique au sein de la mémoire de travail. Notre réflexion théorique fait donc entrer en dialogue des disciplines telles que la linguistique, la psycholinguistique, la didactique, les sciences cognitives. Elle fait également un état des lieux des différents outils numériques, afin de vérifier la compatibilité entre l’enseignement-apprentissage des langues étrangères et la technologie. Notre posture d’enseignant de terrain nous amène à considérer le numérique comme un facteur de motivation engageant fortement les usagers dans l’apprentissage de la langue étrangère. Cela d’autant plus que le public participant à notre étude est généralement peu motivé pour l’apprentissage des langues étrangères. La mise à disposition à des fins pédagogiques de tablettes tactiles, assimilables à des consoles de jeu, est-elle susceptible de les impliquer davantage et de modifier leur représentation de la langue cible? Pour répondre à cette question, nous menons deux expérimentations. Dans un premier temps, nous réalisons un pré-test permettant d’élaborer des hypothèses sur l’engagement cognitif des apprenants dans un environnement numérique. Nous observons l’usage réel et situé de tablettes numériques censées susciter la motivation par leur caractère ludique. Nous examinons également la posture des apprenants et leur capacité à exploiter l’information. A ce stade, la démarche conserve un caractère intuitif et les pratiques observées ne sont pas strictement encadrées. C’est pourquoi, dans un deuxième temps, nous procédons à une expérimentation plus structurée, en meilleure adéquation avec les critères de la recherche-action en langues vivantes étrangères. Nous en détaillons et en analysons les résultats. / The 21st century is commonly known as the “digital age”. Information technology has become a defining component of a new generation of learners, the so-called “digital natives”. Massive investment in digital learning equipment for schools has become a priority to meet the rising demands of society. Digital technologies, initially intended to supplement traditional course book instruction are fast replacing printed books altogether. Instant access to (mostly) authentic language material is a definite asset for developing linguistic skills and implementing the action-based approach prescribed by the Common European Framework of Reference for Languages. However, learner ability does not necessarily match all requirements for processing input from the Internet. Educators who believe in the added value of information technology in language pedagogy should therefore consider more carefully the actual cognitive capacities and cognitive engagement shown by learners when exposed to rich stimulus. For digital devices to become efficient pedagogical tools in language pedagogy, learner ability, teaching strategies, and technological implements must be mutually adjusted and integrated within a consistent framework. Although hypermedia is attractive, non-linear and interactive, it is essential to verify that learners do possess the skills and efficiency to process the information obtained from hypertext documents. Information processing is indeed a major challenge since it involves not only searching and retrieving relevant pieces of information, but requires some cognitive processing—a semantic construction within working memory. The perspective adopted here is thus multidisciplinary, with multiple references to linguistics, education, psycholinguistics, and cognitive science. Also needed are an overview of the various digital devices available and an assessment of their true relevance to language pedagogy. As a professional EFL teacher working in a vocational school, I consider digital technologies to be precious in motivating and engaging learners. The secondary school students involved in the current research study usually show little concern for language learning in the first place. The practical issue can thus be framed as “How can the use of tablets engage them as learners and favourably alter their conceptions of the second language?” To answer this question, a pre-test is set up to form preliminary hypotheses on degrees of cognitive engagement in the language classroom whenever digital devices are put to pedagogical use. The concrete, situated use of tablets is observed, their playful nature and expected contribution to motivation assessed. Learning attitudes and the ability to process information are subject to close scrutiny. At this preliminary stage, methods remain largely intuitive, and the activities or learning processes not strictly characterized. A more structured experiment setup must thus be created, based on more established criteria of action-research, the details and results of which are analysed in this dissertation.