L’étude que nous présentons concerne le rôle du guidage dans l’apprentissage de connaissances scientifiques. En sciences, comme dans d’autres domaines, deux types de connaissances peuvent être distingués (Geary, 2007). D’un côté, les connaissances primaires qui sont des connaissances pour lesquelles l’être humain a évolué ; de l’autre, les connaissances secondaires qui dépendent de la culture et qui évoluent en fonction des changements successifs de cette culture. Selon Geary (2008), alors que les connaissances primaires peuvent être acquises de manière implicite, l’acquisition des connaissances secondaires par les apprenants débutants nécessite un apprentissage explicite et bien souvent, une instruction formelle. Ces deux types de connaissances diffèrent aussi en fonction du coût cognitif qu’ils génèrent. La théorie de la charge cognitive de Sweller (1988) montre que les processus d’apprentissage dépendent de la quantité de ressources attentionnelles en mémoire de travail, requises par l’activité et de la structuration de ces informations dans la mémoire à long terme. Alors que les connaissances primaires peuvent être apprises avec une faible charge cognitive, apprendre les connaissances secondaires va requérir une très grande charge cognitive.Il existe plusieurs formes d’instruction explicite qui ont montré leur efficacité dans les processus d’apprentissage (Renkl, 1997) et notamment le guidage (à partir de l’étude d’exemples travaillés). En effet, le guidage faciliterait l’acquisition des connaissances car il réduirait la quantité de ressources cognitives engagée dans le processus d’apprentissage. Ainsi, le guidage serait inefficace dans le traitement des connaissances primaires et efficace avec celui des connaissances secondaires. Cette recherche a pour but de tester empiriquement cette hypothèse dans le domaine de la biologie. Cinq études ont donc été réalisées auprès de 420 élèves de 6ème de collège. Il s’agissait de comparer trois modalités de guidage : un guidage fort, un guidage adaptatif et une absence de guidage. Les trois premières études concernaient la connaissance primaire de catégorisation impliquée dans la catégorisation du vivant (Etude I), la catégorisation des végétaux (Etude II) et la catégorisation des animaux (Etude III). Ces études, de difficulté croissante, ont montré que le guidage n’était pas efficace dans l’apprentissage de cette connaissance. Les deux autres études concernaient la connaissance secondaire de classification phylogénétique des végétaux (Etude IV) et des animaux (Etude V). Ces études ont montré que, dans ce cas, le guidage était indispensable pour aider les apprenants débutants / This study deals with the role of instruction in learning scientific knowledge. In science as in other fields, two types of knowledge can be distinguished (Geary, 2007): on the one hand, primary knowledge, which is adaptative knowledge; on the other hand, secondary knowledge which depends on culture and which evolves according to this culture’s successive changes. For Geary (2007), primary knowledge can be acquired incidentally or from implicit learning whereas secondary knowledge requires some explicit learning, and often formal instruction.These two knowledge types also differ in terms of cognitive cost. Sweller’s load theory (2008) shows that the learning process depends on the amount of attentional resources of working memory required by the activity and on the structuring of this information in the long-term memory. While primary knowledge can be learnt with a low cognitive load, learning secondary knowledge will require a high cognitive load. Some types of explicit instruction, such as guidance (e.g., worked examples) have proven effective in the learning process (Renkl, 1997). Indeed, instructional guidance would facilitate knowledge acquisition as it would reduce the amount of cognitive resources involved in the learning process. Thus, instructional guidance to help young learners would be ineffective for primary knowledge and necessary for secondary knowledge. This study also aims at examining the effect of instructional guidance in Biology. Five experiments have been conducted with 420 students in 6th grade. Three guidance methods have been compared: strong guidance, adaptative guidance and no guidance. The first three experiments dealt with the primary knowledge of categorization of living species: living beings (Study I), plants (Study II) and animals (Study III). These experiments, of increasing difficulty, have shown that explicit instruction was not effective in learning this type of knowledge.. The other two experiments dealt with the secondary knowledge of the phylogenetic categorization of plants (Study IV) and animals (Study V). These studies have shown that secondary knowledge needs to be taught.

Diese Dissertation enthält eine theoretische und empirische Untersuchung der Gestaltung interaktiver Lernmedien. Interaktivität wird hierbei breit definiert und umfasst sowohl minimal interaktive Benutzersteuerungen, die Lernenden eine Änderung der Darstellung von Lerninhalten erlauben, bis hin zu voll immersiven Umgebungen in der virtuellen Realität. Theoretisch ist die Dissertation auf den Modellen der Cognitive Load Theory und dem Ansatz der Embodied Cognition verankert. Der Ansatz der Cognitive Load Theory rät zu einem Instruktionsdesign bei dem die kognitive Belastung innerhalb einer Lernaufgabe begrenzt wird, jedoch stellt Forschung im Paradigma der Embodied Cognition zumeist die Vorteile von Aktivität für kognitive Prozesse heraus. Die Kapitel dieser Dissertation zielen darauf ab, diese zwei scheinbar widersprüchlichen Ansätze miteinander zu versöhnen. Kapitel 1 bietet einen Überblick über die Inhalte dieser Dissertation und beschreibt die Implikationen der Resultate. In Kapitel 2 wird ein weitreichender Überblick über Forschung zu Embodied Cognition im Bereich des Lernens (auch bekannt als Embodied Learning) präsentiert und eine Taxonomie des Embodied Learning wird beschrieben. Ein wichtiger Beitrag von Kapitel 2 ist die Betrachtung, dass eine Gestaltung von interaktiven Lernumgebungen, die nicht gegen die Grundsätze der Cognitive Load Theory verstoßen, möglich ist. Kapitel 3 leistet eine Reflexion über Detailfragen zur Messung der kognitiven Belastung bei Aufgaben, in denen körperliche Faktoren oder Interaktivität eine Rolle spielen. Eine wichtige Behauptung aus Kapitel 3 besteht darin, dass es Unterschiede hinsichtlich der Passung verschiedener Messinstrumente der kognitiven Belastung für Aufgaben mit einem unterschiedlichen Lehransatz geben könnte. Die Passung verschiedener Fragebogeninstrumente zur Erfassung der (extrinsischen) kognitiven Belastung wird in Kapitel 4 empirisch evaluiert. In zwei Experimenten wurde die extrinsische Belastung (d.h. jene kognitive Belastung, die aus der Umsetzung einer Lernaufgabe herrührt) mit zwei verschiedenen Fragebogeninstrumenten gemessen. Die Versuchspersonen nutzten dabei entweder eine nicht-interaktive oder eine interaktive Implementation von Lernaufgaben zur Anatomie. Bei einem der Experimente stellte sich heraus, dass der gemessene Unterschied in der extrinsischen Belastung zwischen den beiden Versionen bei einem der Fragebögen hoch und bei dem anderen gering ausfiel. Basierend auf diesem Ergebnis wird in dem Kapitel empfohlen, die extrinsische Belastung nicht als einen einheitlichen Begriff, sondern als eine Ansammlung unterschiedlicher Belastungsarten zu betrachten. Darüber hinaus werden in den drei Experimenten in Kapitel 4 Hypothesen hinsichtlich der Gestaltung von Lerntests für Medien aufgestellt, die interaktive Komponenten verwenden. Zum Abschluss zielt Kapitel 5 darauf ab, allgemeine Empfehlungen zur Verbesserung von Forschung im Paradigma der Embodied Cognition zu formulieren und greift dabei auf die Taxonomie aus Kapitel 2 zurück. Insgesamt betrachtet verwendet diese Dissertation die Ansätze der Embodied Cognition und der Cognitive Load Theory um die Theorien gegenseitig anzuregen und um Ansätze zur Überarbeitung beider Theorien zu bieten. / This dissertation presents a theoretical and empirical investigation of the design of interactive learning media. Interactivity is understood in a broad sense, ranging from minimally interactive user controls allowing learners to change the display of learning contents up to fully immersive virtual reality environments. Theoretically, this dissertation is grounded in the models of cognitive load theory and embodied cognition. Cognitive load theory advises instructional designers to limit the cognitive load involved in a learning task, but embodied cognition research usually focuses on the benefits of activity for cognitive processes. The chapters in this dissertation aim to bridge the gap between these two seemingly contradicting approaches. Chapter 1 offers an overview of the contents of this dissertation and describes the implications of the results. In Chapter 2, an extended overview of embodiment research in the field of learning (also referred to as embodied learning) is presented and a taxonomy of embodied learning is outlined. Chapter 2 highlights that there may be ways of designing interactive learning settings while not infringing upon the principles of cognitive load theory. Chapter 3 affords a reflection on the intricacies of measuring cognitive load in tasks that involve embodiment or interactivity. An important assertion of Chapter 3 is that there may be differences in the suitability of different cognitive load measurement instruments for tasks differing in their instructional approach. The suitability of different cognitive load survey instruments for measuring (extraneous) cognitive load is empirically evaluated in Chapter 4. In two experiments, extraneous load (i.e., cognitive load brought about by the implementation of a learning task) was measured using two different cognitive load surveys. Participants either used a noninteractive or an interactive implementation of anatomy learning tasks. In one experiment, the difference in extraneous load between the two versions was high for one survey and low for another survey. Based on this result, the chapter recommends to view extraneous load not as a uniform concept, but rather as a collection of different load types. Moreover, the three studies in Chapter 4 test hypotheses concerning the design of tests for media using interactive learning components. Finally, Chapter 5 intends to formulate general recommendations for advancing embodied cognition research, thereby drawing on the taxonomy of Chapter 2. In sum, this dissertation uses embodied cognition and cognitive load theory to inform each other and to provide a starting point for upgrades for both theories.

While the body of evidence supporting the use of simulation-based learning in the education of health professionals is growing, howor why simulation-based learning works is not yet understood. There is a clear need for evidence, grounded in contemporary educational theory, to clarify the features of simulation instructional design that optimize learning outcomes and efficiency in health care professional students. Cognitive Load Theory (CLT) is a theoretical framework focused on a learner’s working memory capacity. One principle of CLT is example based learning. While this principle has been applied in both traditional classroom and laboratory settings, and has shown positive performance and learning outcomes, example based learning has not yet been applied to the simulation setting. This study had two main objectives: to explore if the example-based learning principle could successfully be applied to the simulation learning environment, and to establish response process validation evidence for a tool designed to measure types of cognitive load. Fifty-eight novice students from nursing, podiatric medicine, physician assistant, physical and occupational therapy programs participated in a blinded randomized control study. The dependent variable was the simulation brief. Participants were randomly assigned to either a traditional brief or a facilitated tutored problem brief. Performance outcomes were measured with verbal communications skill presented in the Introduction, Situation, Background, Assessment, Recommendation (I-SBAR) format. Response process evidence was collected from cognitive interviews of 11 students. Results indicate participation in a tutored problem brief led to statistically significant differences at t(52)=-3.259, p=.002 in verbal communication performance compared to students who participated in a traditional brief. Effect size for this comparison was d=(6.06-4.61)/1.63 = .89 (95% CI 0.32-1.44). Response process evidence demonstrated that additional factors unique to the simulationlearning environment should be accounted for when measuring cognitive load in simulation based learning (SBL). This study suggests that example based learning principles can be successfully applied to SBL and result in positive performance outcomes for health professions students. Additionally, measures of cognitive load do not appear to capture all contribution toload imposed by the simulation environment.

High performance parallel and distributed computing systems are used to solve large, complex, and data parallel scientific applications that require enormous computational power. Data parallel workloads which require performing similar operations on different data objects, are present in a large number of scientific applications, such as N-body simulations and Monte Carlo simulations, and are expressed in the form of loops. Data parallel workloads that lack precedence constraints are called arbitrarily divisible workloads, and are amenable to easy parallelization. Load imbalance that arise from various sources such as application, algorithmic, and systemic characteristics during the execution of scientific applications degrades performance. Scheduling of arbitrarily divisible workloads to address load imbalance in order to obtain better utilization of computing resources is a major area of research. Divisible load theory (DLT) and dynamic loop scheduling (DLS) algorithms are two algorithmic approaches employed in the scheduling of arbitrarily divisible workloads. Despite sharing the same goal of achieving load balancing, the two approaches are fundamentally different. Divisible load theory algorithms are linear, deterministic and platform dependent, whereas dynamic loop scheduling algorithms are probabilistic and platform agnostic. Divisible load theory algorithms have been traditionally used for performance prediction in environments characterized by known or expected variation in the system characteristics at runtime. Dynamic loop scheduling algorithms are designed to simultaneously address all the sources of load imbalance that stochastically arise at runtime from application, algorithmic, and systemic characteristics. In this dissertation, an analysis and performance evaluation of DLT and DLS algorithms are presented in the form of a scalability study and a robustness investigation. The effect of network topology on their performance is studied. A hybrid scheduling approach is also proposed that integrates DLT and DLS algorithms. The hybrid approach combines the strength of DLT and DLS algorithms and improves the performance of the scientific applications running in large scale parallel and distributed computing environments, and delivers performance superior to that which can be obtained by applying DLT algorithms in isolation. The range of conditions for which the hybrid approach is useful is also identified and discussed.

Ownership of mobile devices, such as tablets and smartphones, has quickly risen in the last decade. Unsurprisingly, they are now being integrated into the training and classroom setting. Specifically, the U.S. Army has mapped out a plan in the Army Learning Model of 2015 to utilize mobile devices for training purposes. However, before these tools can be used effectively, it is important to identify how the tablets' unique properties can be leveraged. For this dissertation, the touch interface and the interactivity that tablets afford were investigated using a procedural-motor task. The procedural motor task was the disassembly procedures of a M4 carbine. This research was motivated by cognitive psychology theories, including Cognitive Load Theory and Embodied Cognition. In two experiments, novices learned rifle disassembly procedures in a narrated multimedia presentation presented on a tablet and then were tested on what they learned during the multimedia training involving a virtual rifle by performing a rifle disassembly on a physical rifle, reassembling the rifle, and taking a written recall test about the disassembly procedures. Spatial ability was also considered as a subject variable. Experiment 1 examined two research questions. The primary research question was whether including multiple forms of interactivity in a multimedia presentation resulted in higher learning outcomes. The secondary research question in Experiment 1 was whether dynamic multimedia fostered better learning outcomes than equivalent static multimedia. To examine the effects of dynamism and interactivity on learning, four multimedia conditions of varying levels of interactivity and dynamism were used. One condition was a 2D phase diagram depicting the before and after of the step with no animation or interactivity. Another condition utilized a non-interactive animation in which participants passively watched an animated presentation of the disassembly procedures. A third condition was the interactive animation in which participants could control the pace of the presentation by tapping a button. The last condition was a rifle disassembly simulation in which participants interacted with a virtual rifle to learn the disassembly procedures. A comparison of the conditions by spatial ability yielded the following results. Interactivity, overall, improved outcomes on the performance measures. However, high spatials outperformed low spatials in the simulation condition and the 2D phase diagram condition. High spatials seemed to be able to compensate for low interactivity and dynamism in the 2D phase diagram condition while enhancing their performance in the rifle disassembly simulation condition. In Experiment 2, the touchscreen interface was examined by investigating how gestures and input modality affected learning the disassembly procedures. Experiment 2 had two primary research questions. The first was whether gestures facilitate learning a procedural-motor task through embodied learning. The second was whether direct touch input using resulted in higher learning outcomes than indirect mouse input. To examine the research questions, three different variations of the rifle disassembly simulation were used. One was identical to that of Experiment 1. Another incorporated gestures to initiate the animation whereby participants traced a gesture arrow representing the motion of the component to learn the procedures. The third condition utilized the same interface as the initial rifle disassembly simulation but included "dummy" gesture arrows that displayed only visual information but did not respond to gesture. This condition was included to see the effects (if any) of the gesture arrows in isolation of the gesture component. Furthermore, direct touch input was compared to indirect mouse input. Once again, spatial ability also was considered. Results from Experiment 2 were inconclusive as no significant effects were found. This may have been due to a ceiling effect of performance. However, spatial ability was a significant predictor of performance across all conditions. Overall, the results of the two experiments support the use of multimedia on a tablet to train a procedural-motor task. In line with vision of ALM 2015, the research support incorporating tablets into U.S. Army training curriculum.

Computer-based training has become more prolific as the military and private business enterprises search for more efficient ways to deliver training. However, some methods of computer-based training are not more effective than traditional classroom methods. One technique that may be able to approximate the most effective form of training, one-on-one tutoring, is Adaptive Training (AT). AT techniques use instruction that is tailored to the learner in some way, and can adjust different training parameters such as difficulty, feedback, pace, and delivery mode. There are many ways to adapt training to the learner, and in this study I explored adapting the feedback provided to trainees based on spatial ability in line with Cognitive Load Theory (CLT). In line with the CLT expertise reversal effect literature I hypothesized that for a spatial task, higher ability trainees would perform better when they were given less feedback. Alternately, I hypothesized that lower ability trainees would perform better during training when they were given more support via feedback. This study also compared two different adaptation approaches. The first approach, called the ATI approach, adapts feedback based on a premeasured ability. In this case, it was spatial ability. The second approach, called the Hybrid approach adapts initially based on ability, but then based on performance later in training. I hypothesized that participants who received Hybrid adaptive training would perform better. The study employed a 2(spatial ability; high, low) X 2(feedback; matched, mismatched) X 2 (approach; ATI, Hybrid) between-subjects design in which participants were randomly assigned to one of the eight conditions. Ninety-two participants completed a submarine-based periscope operator task that was visual and spatial in nature. iv The results of the study did not support the use of CLT-derived adaptation based on spatial ability; contrary to what was hypothesized, higher ability participants who received more feedback performed better than those who received less. Similarly, lower ability participants who received less feedback performed better than those who received more. While not significant, results suggested there may be some benefit to using the Hybrid approach, but more research is needed to determine the relative effectiveness of this approach.

Denna studie undersökte hur smartklockors gränssnitt kan utformas för att upplevas användbara samt bidra till minimal kognitiv belastning under fysisk aktivitet. Smartklockors skärmar ställer andra krav på gränssnittet samt medför ett antal utmaningar för hur information kan presenteras på bästa sätt. Utöver detta brukar smartklockor användas i samband med fysisk aktivitet, som ställer höga krav på användarens uppmärksamhet. För att bibehålla hög kvalitet på träningen bör den kognitiva belastningen minska. För att undersöka detta har Cognitive Load Theory applicerats för att kunna optimera arbetsminnets kapacitet. Studien gjorde på personer i 20 års ålder. En kvalitativ datainsamling gjordes genom ett användartest och semistrukturerade intervjuer. Användartestet utfördes på en stationär cykel där deltagarna utförde ett antal uppgifter på en smartklocka. Efteråt skedde semistrukturerade intervjuer. Resultatet analyserades genom kategorisering och jämfördes med tidigare studier. Utifrån de mest förekommande resultaten skapades åtta designförslag. Resultaten visade att eliminering av viss information på gränssnitt kan minska den kognitiva belastningen samt höja användbarheten. Ett konsekvent gränssnitt skulle kunna minimera antal fel samt underlätta för användaren att korrigera fel. En tillämpning av de framtagna designförslagen kan eventuellt minimera den kognitiva belastningen vid interaktion med smartklockors gränssnitt vid fysisk aktivitet. Detta kan underlätta för användaren att fokusera på träningen samt uppleva klockan som mer användbar. / This study examined how an interface for smartwatches can be designed to be perceived as useful and reduce cognitive load during physical activity. Smartwatch screens have other demands on the interface and face several challenges regarding how information can be presented. In addition to this, smartwatches are often used in physical activity, which places high demands on user's attention. To maintain high-quality training, the cognitive load should be reduced. To investigate this, cognitive load theory has been applied to optimize the capacity of the working memory. The study looked at people in their 20s. Qualitative data collection was done through a user test and semi-structured interviews. The user test was done on a stationary bike, the participants performed several tasks on the smartwatch. The participants rode a stationary bicycle while performing tasks on a smartwatch, followed by semistructured interviews. The results were analyzed by thematization and compared with previous studies. Based on the most common results, eight design proposals were created. The results showed that by eliminating unnecessary information, the interface can reduce cognitive load and increase usability. A consistent interface could minimize the number of errors and make it easier for the user to correct errors. An application of the developed design proposals may minimize the external load when interacting with the smartwatch interface during physical activity. This can support users can focus on training and experience the watch as more useful.

Att lösa matematiska uppgifter med hjälp av färdiga lösningsförslag kan ses som ett sätt för elever att anstränga sig mindre och därför lära sig mindre. Det finns dock forskning som har visat att undervisning där eleverna huvudsakligen studerar lösningsförslag har en stor effekt på elevernas förmåga att lösa uppgifter. Den här undervisningsmetoden kallas "the worked example effect" och har fått ett genomslag i den svenska skolan den senaste tiden. Metoden grundas på Swellers teori om kognitiv belastning. Enligt hans och andras forskning kan arbetsminnet delas upp i tre olika typer. Dessa tre, intrinsic cognitive load, germane cognitive load samt extraneous cognitive load, är additiva och stödjer lärande olika mycket. Lösningsförslag har visats minska mängden extraneous cognitive load och öka mängden germane cognitive load. Germane cognitive load har i sin tur visats korrelera med hur väl elever lyckas svara på matematiska uppgifter. Den här studien syftar till att undersöka lösningsförslagens effekt på elevers lärande i den svenska gymnasieskolan. I studien deltog tre skolor, varav en enbart i förstudien, i olika omfattning med totalt 93 elever på natur- och teknikprogrammen i andra årskursen på gymnasiet (17-18-åringar). Elevernas resultat på nationella proven i samtliga obligatoriska matematikkurser användes i undersökningen. Även en enkät genomfördes för att samla in mer information om eleverna och hur de studerade. Studien visar på ett tvetydigt resultat. Över 75% av eleverna anser sig ha ändrat sitt sätt att studera på grund av lösningsförslagen, men om och hur deras lärande har förbättrats, mätt i betyg, är oklart. För en av kurserna i matematik visade studien på ett negativt resultat, men för en annan kurs – positivt. Dessa olika resultat uppkom dessutom på olika skolor. En tolkning av detta resultat kan vara att lösningsförslagens effektivitet är beroende på vilken typ av matematik som ingår i kursen där de används. Det spelar även stor roll hur lösningsförslagen använts av eleverna, något som studien inte undersökt. / To solve mathematical tasks with the help of example solutions can be seen as a way for learners to apply less effort and therefore learn less or perhaps learn in undesirable ways. However, existing international research has indicated that classrooms where pupils mainly study worked examples has had a significant effect on their ability to solve tasks. This method is called "the worked example effect" and has gained increasing support over time. It is based on Sweller’s cognitive load theory. According to his and others’ research the working memory can be divided into three different types. These three, intrinsic cognitive load, germane cognitive load and extraneous cognitive load, are additive and support learning differently. Using worked examples has been shown to reduce the amount of extraneous cognitive load and instead increase the amount of germane cognitive load. Germane cognitive load has in turn been shown to correlate to how well pupils are able to solve mathematical tasks. This thesis aims to study the worked example effect in relation to learning outcome in Swedish secondary school. Three schools took part in the study, of which one only participated in a pilot study. A total of 93 pupils participated, all of them in the science program. Results of the national test for the four latest courses were used in this thesis. The pupils were also asked to fill out a survey with the aim to gather information as to how they used the worked examples when studying. The results of the study are ambiguous. More than 75% of the pupils claim to have changed how they study, but whether their learning has improved, measured in grades, is unclear. The study showed a negative result for one course but a positive for another. These different results originated from different schools. The results of the study may indicate that the worked example effect differs depending on which type of math is being taught. Another important factor is how the worked examples are used by the pupils, something this study has not examined.

The purpose of this study is to explore the possibilities that the highly popular ChatGPT utility might have on education, and more specifically, how it could impact language development in the Swedish EFL classroom. This will be attained through a qualitative method and through multiple observations in real-life situations with students from upper secondary school. Additionally, for a more profound analysis this will then be combined with follow-up interviews with two educating teachers. The observations presented mixed results with some students displaying positive feelings towards ChatGPT while others reacted negatively to its extensive replies to simple inquiries. This proved to be valuable empirical data for this study's analysis. Therefore, this study has concluded that ChatGPT as a didactic tool, should be used in a controlled situation, and explained thoroughly with a specific intentional purpose. In other words, ChatGPT could be used in different teaching situations to enforce students' abilities or to ease the cognitive burden that many students experience when writing texts, as can be explained by cognitive load theory (Sweller, 1988, p. 277). If the messages from the text-generator are clear, informative, personal, and concise it could be used to aid students in their writing process. However, the different complications that arise with the use of ChatGPT including a negative effect on student self-esteem must be considered.

The proliferation of information systems (IS) over the past decades has increased the demand for system authentication. While the majority of system authentications are password-based, it is well documented that passwords have significant limitations. To address this issue, companies have been placing increased requirements on the user to ensure their passwords are more complex and consequently stronger. In addition to meeting a certain complexity threshold, the password must also be changed on a regular basis. As the cognitive load increases on the employees using complex passwords and changing them often, they may have difficulty recalling their passwords. As such, the focus of this experimental study was to determine the effects of raising the cognitive load of the authentication strength for users upon accessing a system via increased strength for passwords requirements. This experimental research uncovered the point at which raising the authentication strength for passwords becomes counterproductive by its impact on end-user performances. To investigate the effects of changing the cognitive load (via different password strength) over time, a quasi-experiment was proposed. Data was collected in an effort to analyze the number of failed operating system (OS) logon attempts, users’ average logon times, average task completion times, and number of requests for assistance (unlock & reset account). Data was also collected for the above relationships when controlled for computer experience, age, and gender. This quasi-experiment included two experimental groups (Group A & B), and a control group (Group C). There was a total of 72 participants from the three groups. Additionally, a pretest-posttest experiment survey was administered before and after the quasi-experiment. Such assessment was done in an effort to see if user’s perceptions of password use would be changed by participating in this experimental study. The results indicated a significant difference between the user’s perceptions about passwords before and after the quasi-experiment. The Multivariate Analysis of Variance (MANOVA) and Multivariate Analysis of Covariate (MANCOVA) tests were conducted. The results revealed a significance difference on the number of failed logon attempts, average logon times, average task completion, and amount of request for assistance between the three groups (two treatment groups & the control group). However, no significant differences were observed when controlling for computer experience, age, and gender. This research study contributed to the body of knowledge and has implications for industry as well as for further study in the information systems domain. It contributed by giving insight into the point at which an increase of the cognitive load (via different password strengths) become counterproductive to the organization by causing an increase in number of failed OS logon attempts, users' average logon times, average task completion times, and number of requests for assistance (unlock and reset account). Future studies may be conducted in the industry as results by differ from college students.