Metacognition | EPFL Graph Search

Metacognition is an awareness of one's thought processes and an understanding of the patterns behind them. The term comes from the root word meta, meaning "beyond", or "on top of". Metacognition can take many forms, such as reflecting on one's ways of thinking and knowing when and how to use particular strategies for problem-solving. There are generally two components of metacognition: (1) knowledge about cognition and (2) regulation of cognition. A metacognitive model differs from other scientific models in that the creator of the model is per definition also enclosed within it. Scientific models are often prone to distancing the observer from the object or field of study whereas a metacognitive model in general tries to include the observer in the model. Metamemory, defined as knowing about memory and mnemonic strategies, is an especially important form of metacognition. Academic research on metacognitive processing across cultures is in the early stages, but

Neural correlates of performance monitoring during discrete and continuous tasks

Michael Eric Anthony Pereira

Monitoring our actions is a key function of the brain for adaptive and successful behavior. Actions can be discrete such as when pressing a button, or continuous, such as driving a car. Moreover, we evaluate our actions as correct or erroneous (performance monitoring) and this appraisal of performance comes with various levels of confidence (metacognition). However, studies of performance monitoring have focused on discrete actions and are mostly agnostic to metacognitive judgments. The objective of this thesis was to extend the study of performance monitoring to more ecological conditions, in which monitoring occurs during continuous motor tasks under various degrees of error and confidence level. We first investigated the role of actions in performance monitoring together with metacognitive judgments, using simultaneous EEG and fMRI recordings. To dissociate the role of motor actions, we designed an experimental paradigm in which subjects had to rate their confidence level about an action that they had either performed themselves (a button press) based on a decision or passively observed (a virtual hand displayed). We found correlates of confidence in both condition, in the EEG and in the supplementary motor area (SMA). Furthermore, we found that subject showed better metacognitive performances when they were the agents of the action. This difference was further emphasized for subjects that showed higher activations of a network previously linked to motor inhibition and comprising the pre-SMA and inferior frontal gyrus (IFG). Our results imply that the SMA plays a primary role in the monitoring of performance, irrespectively of a commitment to a decision and the resulting action. Our findings also suggest that the additional neural processes leading to decisions and actions can inform the metacognitive judgments. In the following chapters, we ask whether electrophysiological correlates of performance monitoring can be found in less experimentally constrained paradigms for which motor output continuous unfolds and visual feedback is not segregated into discrete events. By decomposing the unfolding hand kinematics during a visuo-motor tracking task into periodic acceleration pulses âhenceforth referred to as sub-movements, we found three electrophysiological markers that could possibly be linked to performance monitoring. Firstly, we found an ERP in the SMA, time-locked to sub-movements which encoded the deviation of the hand, 110 ms before. Secondly, we found high-gamma activity in the ACC and SMA of epileptic patients, that was phase-locked to sub-movements. Thirdly, we found a transient modulation of mu oscillations over the ipsilateral sensorimotor cortices that depended on sub-movement amplitude. Altogether, these results provide a strong contribution in the understanding of the neurophysiological processes underlying performance monitoring. Our work proposes a methodological framework to study electrophysiological correlates of performance monitoring in less controlled paradigms during which continuous visual feedback has to be constantly integrated into motor corrections. In the conclusion chapter, we propose a way of extending current models of performance monitoring and decision making to explain the findings of this thesis by considering continuous motor tasks as a succession of decision making processes under time pressure and uncertainty.

EPFL2018

A generic methodology and a digital twin for zero defect manufacturing (ZDM) performance mapping towards design for ZDM

Foivos Psarommatis Giannakopoulos

Over recent years, the manufacturing industry has seen constant growth and change. From one side, it has been affected by the fourth industrial revolution (Industry 4.0). From the other side, it has had to enhance its ability to meet higher customer expectations, such as producing more customized products in a shorter time. In the contemporary competitive market of manufacturing, quality is a criterion of primary importance for winning market share. Quality improvement must be coupled with a concern for high performance. One of the most promising concepts for quality control and improvement is called zero defect manufacturing (ZDM), which utilizes the benefits of Industry 4.0 technologies. ZDM imposes the rule that any event in the production process should have a counter-action to mitigate it. In light of this, the current research developed a methodology the manufacturer can use to correctly select or design appropriate ZDM strategies and equipment to implement at each manufacturing stage. This methodology consists of several steps. The first step is to conduct several simulations using a dynamic scheduling tool with specific data sets to develop a digital twin (DT). The data sets are created using the Taguchi design of experiments methodology. The DT model is created for use in predicting the results of the developed scheduling tool without actually using said tool. Using the DT, multiple ZDM parameter-combination sets can be created and plugged into the model. This process generates ZDM performance maps that show the effect of each ZDM strategy at each manufacturing stage under different control parameters. These maps are intended to provide information for comparing different ZDM-oriented equipment to help manufacturers reach a final decision on correct and efficient ZDM implementation or to assist in the design phase of a ZDM strategy implementation.

2021