Refine
Has Fulltext
- yes (340) (remove)
Year of publication
- 2023 (340) (remove)
Document Type
- Doctoral Thesis (181)
- Article (55)
- Postprint (35)
- Monograph/Edited Volume (19)
- Master's Thesis (13)
- Working Paper (11)
- Part of Periodical (8)
- Bachelor Thesis (5)
- Report (3)
- Conference Proceeding (2)
Is part of the Bibliography
- yes (340) (remove)
Keywords
- Reflexion (10)
- Digitalisierung (9)
- climate change (9)
- Klimawandel (8)
- machine learning (8)
- digitalization (7)
- digital education (5)
- e-learning (5)
- teacher training (5)
- Lehrkräftebildung (4)
Institute
- Extern (49)
- Institut für Biochemie und Biologie (30)
- Institut für Geowissenschaften (29)
- Institut für Physik und Astronomie (28)
- Hasso-Plattner-Institut für Digital Engineering GmbH (27)
- Zentrum für Lehrerbildung und Bildungsforschung (ZeLB) (27)
- Institut für Chemie (22)
- Bürgerliches Recht (15)
- Institut für Umweltwissenschaften und Geographie (14)
- Department Erziehungswissenschaft (11)
Literarische Grammatik
(2023)
Dieser Band versammelt neun Beiträge mit dem Ziel, Sprach- und Literaturwissenschaft aufeinander zu beziehen: Literatur grammatisch zu betrachten und Grammatik für Literatur (neu) zu denken. Jeder Beitrag nimmt mindestens einen grammatischen und einen literarischen Gegenstand zum Ausgangspunkt. Dabei ist die Bandbreite groß; sie reicht von Bodo Kirchhoffs Roman ‚Dämmer und Aufruhr‘ über die Kurzgeschichte ‚Das Brot‘ von Wolfgang Borchert bis hin zu Marion Poschmanns Gedichtzyklus ‚Kindergarten Lichtenberg‘ und deckt unterschiedlichste sprachliche Bereiche wie Tempus, semantische Rollen, Interpunktionszeichen oder Metaphern ab.
Ist es in der Schule geradezu erwünscht, Grammatik und Literatur integrativ zu unterrichten, verfolgen sie als universitäre Disziplinen oft ganz unterschiedliche Fragestellungen an verschiedenen Sprachwerken. Vor diesem Hintergrund ist dieser Band ein interdisziplinärer Versuch, Anregungen und neue Perspektiven für schulische wie universitäre Bildungskontexte zu geben.
Throughout the last ~3 million years, the Earth's climate system was characterised by cycles of glacial and interglacial periods. The current warm period, the Holocene, is comparably stable and stands out from this long-term cyclicality. However, since the industrial revolution, the climate has been increasingly affected by a human-induced increase in greenhouse gas concentrations. While instrumental observations are used to describe changes over the past ~200 years, indirect observations via proxy data are the main source of information beyond this instrumental era. These data are indicators of past climatic conditions, stored in palaeoclimate archives around the Earth. The proxy signal is affected by processes independent of the prevailing climatic conditions. In particular, for sedimentary archives such as marine sediments and polar ice sheets, material may be redistributed during or after the initial deposition and subsequent formation of the archive. This leads to noise in the records challenging reliable reconstructions on local or short time scales. This dissertation characterises the initial deposition of the climatic signal and quantifies the resulting archive-internal heterogeneity and its influence on the observed proxy signal to improve the representativity and interpretation of climate reconstructions from marine sediments and ice cores.
To this end, the horizontal and vertical variation in radiocarbon content of a box-core from the South China Sea is investigated. The three-dimensional resolution is used to quantify the true uncertainty in radiocarbon age estimates from planktonic foraminifera with an extensive sampling scheme, including different sample volumes and replicated measurements of batches of small and large numbers of specimen. An assessment on the variability stemming from sediment mixing by benthic organisms reveals strong internal heterogeneity. Hence, sediment mixing leads to substantial time uncertainty of proxy-based reconstructions with error terms two to five times larger than previously assumed.
A second three-dimensional analysis of the upper snowpack provides insights into the heterogeneous signal deposition and imprint in snow and firn. A new study design which combines a structure-from-motion photogrammetry approach with two-dimensional isotopic data is performed at a study site in the accumulation zone of the Greenland Ice Sheet. The photogrammetry method reveals an intermittent character of snowfall, a layer-wise snow deposition with substantial contributions by wind-driven erosion and redistribution to the final spatially variable accumulation and illustrated the evolution of stratigraphic noise at the surface. The isotopic data show the preservation of stratigraphic noise within the upper firn column, leading to a spatially variable climate signal imprint and heterogeneous layer thicknesses. Additional post-depositional modifications due to snow-air exchange are also investigated, but without a conclusive quantification of the contribution to the final isotopic signature.
Finally, this characterisation and quantification of the complex signal formation in marine sediments and polar ice contributes to a better understanding of the signal content in proxy data which is needed to assess the natural climate variability during the Holocene.
Portal Wissen = Lernen
(2023)
Uns lernend zu verändern, ist eine der wichtigsten Eigenschaften, die wir Menschen haben. Wir werden geboren und können – scheinbar – nichts, müssen uns alles erst erschließen, abschauen, aneignen: greifen und laufen, essen und sprechen. Natürlich auch lesen und rechnen. Inzwischen wissen wir: Damit werden wir nie fertig. Im besten Fall lernen wir ein Leben lang. Hören wir damit auf, schadet es uns. „Es ist keine Schande, nichts zu wissen, wohl aber, nichts lernen zu wollen“, meinte vor über 2.400 Jahren schon der griechische Philosoph Platon.
Auch als Menschheit sind wir lernfähig, gelangten dank immer mehr Wissen über die Welt um uns herum aus der Steinzeit ins digitale Zeitalter. Dass auch dieser Fortschritt keine Ziellinie ist, sondern wir nach wie vor einen weiten Weg vor uns haben, zeigen der menschengemachte Klimawandel – und vor allem die Unfähigkeit, als globale Gemeinschaft das, was uns die Forschung lehrt, in entsprechendes Handeln zu übersetzen. Bleibt zu hoffen, dass wir das auch noch begreifen.
Was wir in der intensiven Diskussion über die vielschichtigen Ebenen des Lernens gern übersehen: Wir sind keineswegs die einzig Lernenden. Viele, wenn nicht alle Lebewesen auf der Erde lernen, manche zielstrebiger und komplexer, kognitiver, als andere. Und seit einiger Zeit sind auch Maschinen in der Lage, mehr oder weniger selbstständig zu lernen. Künstliche Intelligenz lässt grüßen.
Lernen kann in seiner Bedeutung für den Menschen kaum überschätzt werden. Das hat auch die Wissenschaft begriffen und die Lernprozesse und -bedingungen in nahezu allen Zusammenhängen für sich entdeckt, egal, ob es um unsere eigenen geht oder solche um uns herum. Einigen davon sind wir für die aktuelle Ausgabe der „Portal Wissen“ nachgegangen.
So erforscht die Neurowissenschaftlerin Milena Rabovsky, wie unser Hirn gesprochene Sprache vorhersagt – und dabei aus seinen Fehlern lernt –, während die Psycholinguistin Natalie Boll-Avetisyan eine Box entwickelt hat, mit der sich schon bei kleinen Kindern Störungen beim Sprachenlernen entdecken lassen. Die Verhaltensbiologinnen Jana Eccard und Valeria Mazza haben das Verhalten von kleinen Nagetieren untersucht und dabei nicht nur festgestellt, dass sie sehr unterschiedliche Persönlichkeiten ausbilden, sondern auch beschrieben, wie sie lernen, diese an wechselnde Umweltbedingungen anzupassen. Die Bildungsforscherin Katharina Scheiter erklärt, wie die Möglichkeiten der Digitalisierung unser Lernen verändern – und wie nicht. Der Politikwissenschaftler Fabian Schuppert und die Verwaltungsexpertin Sabine Kuhlmann wiederum analysieren die Klimapolitiken von Millionenstädten überall auf der Welt – und dabei vor allem die Art und Weise, wie die Bevölkerung einbezogen wird –, damit die Metropolregion Berlin von diesen Strategien profitieren kann. Und der Computerlinguist David Schlangen geht der Frage nach, was Maschinen lernen müssen, damit unsere Kommunikation mit ihnen noch besser funktioniert.
Da Forschung letztlich immer ein Lernprozess ist, der danach strebt, etwas zu verstehen, was bislang noch unbekannt ist, stehen dieses Mal ohnehin alle Texte irgendwie unter dem „Stern“ des Titelthemas: Es geht darum, wie wir Millionen Jahre alte Korallen als Klimaarchive lesen können, was die Geschichte der vergangenen Jahrhunderte uns über „Militärische Gewaltkulturen“ verrät und die Frage, welche Lehren wir aus Naturgefahren für die Zukunft ziehen sollten.
Wir haben mit einer Juristin gesprochen, die über den Tellerrand der Universität blickt und Recht für jedermann verständlich machen will, und mit einem Philosophen, der untersucht, warum „eine Meinung haben“ heute etwas anderes bedeutet als vor 100 Jahren. Wir berichten von „smarter DNA“ und der KIgestützten Genomanalyse, die beide die Gesundheitsversorgung nachhaltig verändern können. Außerdem geht es um das Berufsbild „YouTuber*in“, ein Start-up, das eine App entwickelt hat, dank der Paare spielerisch ihre Liebe vertiefen können, und die Frage, wie sich unser mentales Lexikon erforschen lässt. Wir sprechen über minor cosmopolitanisms, Wildtiermanagement in Afrika und Wasserstoff als Energiequelle der Zukunft. Wenn Sie hier fertig sind, haben Sie was gelernt. Versprochen! Viel Vergnügen!
Portal Wissen = Learning
(2023)
Changing through learning is one of the most important characteristics we humans have. We are born and can – it seems – do nothing. We have to comprehend, copy, and acquire everything: grasping and walking, eating and speaking. Of course, we also have to read and do number work. In the meantime, we know: We will never be able to finish this. At best, we learn for a lifetime. If we stop, it harms us. The Greek philosopher Plato said more than 2,400 years ago, “There is no shame in not knowing something. The shame is in not being willing to learn.”
As humans we are also capable of learning; thanks to more and more knowledge about the world around us, we have moved from the Stone Age into the digital age. That this development is not a finish line either, but that we still have a long way to go, is shown by man-made climate change – and above all by our inability as a global community to translate what research teaches us into appropriate actions. Let us dare to hope that we also comprehend this.
What we tend to ignore in the intensive discussion about the multi-layered levels of learning: We are by no means the only learners. Many, if not all, living beings on our planet learn, some more in a more purposeful and complex and more cognitive way than others. And for some time now, machines have also been able to learn more or less independently. Artificial intelligence sends its regards.
The significance of learning for human beings can hardly be overestimated. Science has also understood this and has discovered the learning processes and conditions in almost all contexts for itself, no matter whether it is about our own learning processes and conditions or those around us. We have investigated some of these for the current issue of “Portal Wissen”.
Psycholinguist Natalie Boll-Avetisyan has developed a box that can be used to detect language learning disorders already in young children. The behavioral biologists Jana Eccard and Valeria Mazza investigated the behavior of small rodents and found out that they do not only develop different personality traits but they also described how they learn to adapt them different environmental conditions. Computer linguist David Schlangen examines the question what machines have to learn so that our communication with them works even better.
Since research is ultimately always a learning process that strives to understand something yet unknown, this time all texts are somehow along the motto of the title theme: It is about what the history of past centuries reveals about “military cultures of violence” and the question of what lessons we should learn from natural hazards for the future.
We talked with a legal scholar who looks beyond the university’s backyard and wants to make law comprehensible to everyone. We also talked with a philosopher who analyzes why “having an opinion” means something different today than 100 years ago. We report about an AI-based genome analysis that can change healthcare sustainably. Furthermore, it is about the job profile “YouTuber”, minor cosmopolitanisms, and wildlife management in Africa. When you have finished reading, you will have learnt something. Promised! Enjoy your read!
Diese Masterarbeit zielt darauf ab, exemplarisch an zoologischen Gärten für das politische Spannungsverhältnis zwischen Mensch und Tier zu sensibilisieren sowie die damit einhergehenden Aushandlungsprozesse auf individueller bzw. gesamtgesellschaftlicher Ebene didaktisch anschlussfähig zu machen. Nach einer kurzen begrifflichen Einführung der titelgebenden Termini werden in diesem Sinne vier verschiedene Ausdrucksformen ambivalenter Mensch-Tier-Beziehungen erörtert: die Entwicklungsgeschichte und Architektur sowie die Artenschutz- bzw. Bildungsleistungen der Zoos. Dabei wird der historisch vorbelastete Balanceakt zoologischer Gärten deutlich, in Gegenwart und Zukunft menschliche und tierliche Interessen glaubhaft in Einklang bringen zu müssen. Als Grundübel dieses Dilemmas wird wiederum der menschliche Anspruch identifiziert, Naturzustände vor dem Hintergrund eines fragwürdigen Legitimationsnarratives kulturell nachstellen zu wollen.
Außerdem entfaltet der Autor die These, dass der Zoo gerade durch die ihn prägenden Ambivalenzen gegenüber anderen Problembereichen der Mensch-Tier-Beziehungen an Kontroversität gewinnt und somit prädestiniert ist, um als politikdidaktische Reibungsfläche zeitgemäßer Mensch-Tier-Beziehungen zu fungieren. Dementsprechend werden abschließend Zugänge vorgeschlagen, um den Zoo als außerschulischen politischen Lernort vor dem Hintergrund vielfältiger Streitfragen erkunden und produktiv erörtern zu können.
Indem Schülerinnen und Schüler demnach die Wert- und Zweckrationalität der Zoos auf den Prüfstand stellen, werden sie dazu angeregt, sich selbst- und gesellschaftskritisch mit dem politischen Verhältnis zwischen Tieren und Menschen auseinanderzusetzen. Die dabei exemplarisch am Zoo gewonnenen Erkenntnisse und Überzeugungen lassen sich in Bezug auf die ebenso drängende wie polarisierende Tierfrage abstrahieren. Durch den somit geschaffenen Orientierungsrahmen werden die Lernenden nicht zuletzt in die Lage versetzt, ihre gereiften Vorstellungen von einem angemessenen Umgang mit (nichtmenschlichen) Tieren öffentlich zu vertreten.
Evaluation of nitrogen dynamics in high-order streams and rivers based on high-frequency monitoring
(2023)
Nutrient storage, transform and transport are important processes for achieving environmental and ecological health, as well as conducting water management plans. Nitrogen is one of the most noticeable elements due to its impacts on tremendous consequences of eutrophication in aquatic systems. Among all nitrogen components, researches on nitrate are blooming because of widespread deployments of in-situ high-frequency sensors. Monitoring and studying nitrate can become a paradigm for any other reactive substances that may damage environmental conditions and cause economic losses.
Identifying nitrate storage and its transport within a catchment are inspiring to the management of agricultural activities and municipal planning. Storm events are periods when hydrological dynamics activate the exchange between nitrate storage and flow pathways. In this dissertation, long-term high-frequency monitoring data at three gauging stations in the Selke river were used to quantify event-scale nitrate concentration-discharge (C-Q) hysteretic relationships. The Selke catchment is characterized into three nested subcatchments by heterogeneous physiographic conditions and land use. With quantified hysteresis indices, impacts of seasonality and landscape gradients on C-Q relationships are explored. For example, arable area has deep nitrate legacy and can be activated with high intensity precipitation during wetting/wet periods (i.e., the strong hydrological connectivity). Hence, specific shapes of C-Q relationships in river networks can identify targeted locations and periods for agricultural management actions within the catchment to decrease nitrate output into downstream aquatic systems like the ocean.
The capacity of streams for removing nitrate is of both scientific and social interest, which makes the quantification motivated. Although measurements of nitrate dynamics are advanced compared to other substances, the methodology to directly quantify nitrate uptake pathways is still limited spatiotemporally. The major problem is the complex convolution of hydrological and biogeochemical processes, which limits in-situ measurements (e.g., isotope addition) usually to small streams with steady flow conditions. This makes the extrapolation of nitrate dynamics to large streams highly uncertain. Hence, understanding of in-stream nitrate dynamic in large rivers is still necessary. High-frequency monitoring of nitrate mass balance between upstream and downstream measurement sites can quantitatively disentangle multi-path nitrate uptake dynamics at the reach scale (3-8 km). In this dissertation, we conducted this approach in large stream reaches with varying hydro-morphological and environmental conditions for several periods, confirming its success in disentangling nitrate uptake pathways and their temporal dynamics. Net nitrate uptake, autotrophic assimilation and heterotrophic uptake were disentangled, as well as their various diel and seasonal patterns. Natural streams generally can remove more nitrate under similar environmental conditions and heterotrophic uptake becomes dominant during post-wet seasons. Such two-station monitoring provided novel insights into reach-scale nitrate uptake processes in large streams.
Long-term in-stream nitrate dynamics can also be evaluated with the application of water quality model. This is among the first time to use a data-model fusion approach to upscale the two-station methodology in large-streams with complex flow dynamics under long-term high-frequency monitoring, assessing the in-stream nitrate retention and its responses to drought disturbances from seasonal to sub-daily scale. Nitrate retention (both net uptake and net release) exhibited substantial seasonality, which also differed in the investigated normal and drought years. In the normal years, winter and early spring seasons exhibited extensive net releases, then general net uptake occurred after the annual high-flow season at later spring and early summer with autotrophic processes dominating and during later summer-autumn low-flow periods with heterotrophy-characteristics predominating. Net nitrate release occurred since late autumn until the next early spring. In the drought years, the late-autumn net releases were not so consistently persisted as in the normal years and the predominance of autotrophic processes occurred across seasons. Aforementioned comprehensive results of nitrate dynamics on stream scale facilitate the understanding of instream processes, as well as raise the importance of scientific monitoring schemes for hydrology and water quality parameters.
Records from ocean bottom seismometers (OBSs) are highly contaminated by noise, which is much stronger
compared to data from most land stations, especially on the horizontal components. As a consequence, the high energy of the oceanic noise at frequencies below 1 Hz considerably complicates the analysis of the teleseismic earthquake signals recorded by OBSs.
Previous studies suggested different approaches to remove low-frequency noises from OBS recordings but mainly focused on the vertical component. The records of horizontal components, which are crucial for the application of many methods in passive seismological analysis of body and surface waves, could not be much improved in the teleseismic frequency band. Here we introduce a noise reduction method, which is derived from the harmonic–percussive separation algorithms used in Zali et al. (2021), in order to separate long-lasting narrowband signals from broadband transients in the OBS signal. This leads to significant noise reduction of OBS records on both the vertical and horizontal components and increases the earthquake signal-to-noise ratio (SNR) without distortion of the broadband earthquake waveforms. This is demonstrated through tests with synthetic data. Both SNR and cross-correlation coefficients showed significant improvements for different realistic noise realizations. The application of denoised signals in surface wave analysis and receiver functions is discussed through tests with synthetic and real data.
Volcanoes are one of the Earth’s most dynamic zones and responsible for many changes in our planet. Volcano seismology aims to provide an understanding of the physical processes in volcanic systems and anticipate the style and timing of eruptions by analyzing the seismic records. Volcanic tremor signals are usually observed in the seismic records before or during volcanic eruptions. Their analysis contributes to evaluate the evolving volcanic activity and potentially predict eruptions. Years of continuous seismic monitoring now provide useful information for operational eruption forecasting. The continuously growing amount of seismic recordings, however, poses a challenge for analysis, information extraction, and interpretation, to support timely decision making during volcanic crises. Furthermore, the complexity of eruption processes and precursory activities makes the analysis challenging.
A challenge in studying seismic signals of volcanic origin is the coexistence of transient signal swarms and long-lasting volcanic tremor signals. Separating transient events from volcanic tremors can, therefore, contribute to improving our understanding of the underlying physical processes. Some similar issues (data reduction, source separation, extraction, and classification) are addressed in the context of music information retrieval (MIR). The signal characteristics of acoustic and seismic recordings comprise a number of similarities. This thesis is going beyond classical signal analysis techniques usually employed in seismology by exploiting similarities of seismic and acoustic signals and building the information retrieval strategy on the expertise developed in the field of MIR.
First, inspired by the idea of harmonic–percussive separation (HPS) in musical signal processing, I have developed a method to extract harmonic volcanic tremor signals and to detect transient events from seismic recordings. This provides a clean tremor signal suitable for tremor investigation along with a characteristic function suitable for earthquake detection. Second, using HPS algorithms, I have developed a noise reduction technique for seismic signals. This method is especially useful for denoising ocean bottom seismometers, which are highly contaminated by noise. The advantage of this method compared to other denoising techniques is that it doesn’t introduce distortion to the broadband earthquake waveforms, which makes it reliable for different applications in passive seismological analysis. Third, to address the challenge of extracting information from high-dimensional data and investigating the complex eruptive phases, I have developed an advanced machine learning model that results in a comprehensive signal processing scheme for volcanic tremors. Using this method seismic signatures of major eruptive phases can be automatically detected. This helps to provide a chronology of the volcanic system. Also, this model is capable to detect weak precursory volcanic tremors prior to the eruption, which could be used as an indicator of imminent eruptive activity. The extracted patterns of seismicity and their temporal variations finally provide an explanation for the transition mechanism between eruptive phases.
In this work, the role of the TusA protein was investigated for the cell functionality and FtsZ ring assembly in Escherichia coli. TusA is the tRNA-2-thiouridine synthase that acts as a sulfur transferase in tRNA thiolation for the formation of 2-thiouridine at the position 34 (wobble base) of tRNALys, tRNAGlu and tRNAGln. It binds the persulfide form of sulfur and transfers it to further proteins during mnm5s2U tRNA modification at wobble position and for Moco biosynthesis. With this thiomodification of tRNA, the ribosome binding is more efficient and frameshifting is averted during the protein translation. Previous studies have revealed an essential role of TusA in bacterial cell physiology since deletion of the tusA gene resulted in retarded growth and filamentous cells during the exponential growth phase in a rich medium which suddenly disappeared during the stationary phase. This indicates a problem in the cell division process. Therefore the focus of this work was to investigate the role of TusA for cell functionality and FtsZ ring formation and thus the cell separation.
The reason behind the filamentous growth of the tusA mutant strain was investigated by growth and morphological analyses. ΔtusA cells showed a retarded growth during the exponential phase compared to the WT strain. Also, morphological analysis of ΔtusA cells confirmed the filamentous cell shape. The growth and cell division defects in ΔtusA indicated a defect in FtsZ protein as a key player of cell division. The microscopic investigation revealed that filamentous ΔtusA cells possessed multiple DNA parts arranged next to each other. This suggested that although the DNA replication occurred correctly, there was a defect in the step where FtsZ should act; probably FtsZ is unable to assemble to the ring structure or the assembled ring is not able to constrict. All tested mutant strains (ΔtusD, ΔtusE and ΔmnmA) involved in the mnm5s2U34 tRNA modification pathway shared the similar retarded growth and filamentous cell shape like ΔtusA strain. Thus, the cell division defect arises from a defect in mnm5s2U34 tRNA thiolation.
Since the FtsZ ring formation was supposed to be defective in filaments, a possible intracellular interaction of TusA and FtsZ was examined by fluorescent (EGFP and mCherry) fusion proteins expression and FRET. FtsZ expressing tusA mutant (DE3) cells showed a red mCherry signal at the cell poles, indicating that FtsZ is still in the assembling phase. Interestingly, the cellular region of EGFP-TusA fusion protein expressed in ΔtusA (DE3) was conspicuous; the EGFP signal was spread throughout the whole cell and, in addition, a slight accumulation of the EGFP-TusA fluorescence was detectable at the cell poles, the same part of the cell as for mCherry-FtsZ. Thus, this strongly suggested an interaction of TusA and FtsZ.
Furthermore, the cellular FtsZ and Fis concentrations, and their change during different growth phases were determined via immunoblotting. All tested deletion strains of mnm5s2U34 tRNA modification show high cellular FtsZ and Fis levels in the exponential phase, shifting to the later growth phases. This shift reflects the retarded growth, whereby the deletion strains reach later the exponential phase. Conclusively, the growth and cell division defect, and thus the formation of filaments, is most likely caused by changes in the cellular FtsZ and Fis concentrations.
Finally, the translation efficiencies of certain proteins (RpoS, Fur, Fis and mFis) in tusA mutant and in additional gene deletion strains were studied whether they were affected by using unmodified U34 tRNAs of Lys, Glu and Gln. The translation efficiency is decreased in mnm5s2U34 tRNA modification-impaired strains in addition to their existing growth and cell division defect due to the elimination of these three amino acids. Finally, these results confirm and reinforce the importance of Lys, Glu and Gln and the mnm5s2U34 tRNA thiolation for efficient protein translation. Thus, these findings verify that the translation of fur, fis and rpoS is regulated by mnm5s2U34 tRNA modifications, which is growth phase-dependent.
In total, this work showed the importance of the role of TusA for bacterial cell functionality and physiology. The deletion of the tusA gene disrupted a complex regulatory network within the cell, that most influenced by the decreased translation of Fis and RpoS, caused by the absence of mnm5s2U34 tRNA modifications. The disruption of RpoS and Fis cellular network influences in turn the cellular FtsZ level in the early exponential phase. Finally, the reduced FtsZ concentration leads to elongated, filamentous E. coli cells, which are unable to divide.
Successful sentence comprehension requires the comprehender to correctly figure out who did what to whom. For example, in the sentence John kicked the ball, the comprehender has to figure out who did the action of kicking and what was being kicked. This process of identifying and connecting the syntactically-related words in a sentence is called dependency completion. What are the cognitive constraints that determine dependency completion? A widely-accepted theory is cue-based retrieval. The theory maintains that dependency completion is driven by a content-addressable search for the co-dependents in memory. The cue-based retrieval explains a wide range of empirical data from several constructions including subject-verb agreement, subject-verb non-agreement, plausibility mismatch configurations, and negative polarity items.
However, there are two major empirical challenges to the theory: (i) Grammatical sentences’ data from subject-verb number agreement dependencies, where the theory predicts a slowdown at the verb in sentences like the key to the cabinet was rusty compared to the key to the cabinets was rusty, but the data are inconsistent with this prediction; and, (ii) Data from antecedent-reflexive dependencies, where a facilitation in reading times is predicted at the reflexive in the bodybuilder who worked with the trainers injured themselves vs. the bodybuilder who worked with the trainer injured themselves, but the data do not show a facilitatory effect.
The work presented in this dissertation is dedicated to building a more general theory of dependency completion that can account for the above two datasets without losing the original empirical coverage of the cue-based retrieval assumption. In two journal articles, I present computational modeling work that addresses the above two empirical challenges.
To explain the grammatical sentences’ data from subject-verb number agreement dependencies, I propose a new model that assumes that the cue-based retrieval operates on a probabilistically distorted representation of nouns in memory (Article I). This hybrid distortion-plus-retrieval model was compared against the existing candidate models using data from 17 studies on subject-verb number agreement in 4 languages. I find that the hybrid model outperforms the existing models of number agreement processing suggesting that the cue-based retrieval theory must incorporate a feature distortion assumption.
To account for the absence of facilitatory effect in antecedent-reflexive dependencies, I propose an individual difference model, which was built within the cue-based retrieval framework (Article II). The model assumes that individuals may differ in how strongly they weigh a syntactic cue over a number cue. The model was fitted to data from two studies on antecedent-reflexive dependencies, and the participant-level cue-weighting was estimated. We find that one-fourth of the participants, in both studies, weigh the syntactic cue higher than the number cue in processing reflexive dependencies and the remaining participants weigh the two cues equally. The result indicates that the absence of predicted facilitatory effect at the level of grouped data is driven by some, not all, participants who weigh syntactic cues higher than the number cue. More generally, the result demonstrates that the assumption of differential cue weighting is important for a theory of dependency completion processes. This differential cue weighting idea was independently supported by a modeling study on subject-verb non-agreement dependencies (Article III).
Overall, the cue-based retrieval, which is a general theory of dependency completion, needs to incorporate two new assumptions: (i) the nouns stored in memory can undergo probabilistic feature distortion, and (ii) the linguistic cues used for retrieval can be weighted differentially. This is the cumulative result of the modeling work presented in this dissertation.
The dissertation makes an important theoretical contribution: Sentence comprehension in humans is driven by a mechanism that assumes cue-based retrieval, probabilistic feature distortion, and differential cue weighting. This insight is theoretically important because there is some independent support for these three assumptions in sentence processing and the broader memory literature. The modeling work presented here is also methodologically important because for the first time, it demonstrates (i) how the complex models of sentence processing can be evaluated using data from multiple studies simultaneously, without oversimplifying the models, and (ii) how the inferences drawn from the individual-level behavior can be used in theory development.