Discovery and Temporal Analysis of MOOC Study Patterns

The large-scale and granular interaction data collected in online learning platforms such as massive open online courses (MOOCs) provide unique opportunities to better understand individuals’ learning processes and could facilitate the design of personalized and more effective support mechanisms for learners. In this paper, we present two different methods of extracting study patterns from activity sequences. Unlike most of the previous works, with post hoc analysis of activity patterns, our proposed methods could be deployed during the course and enable the learners to receive real-time support and feedback. In the first method, following a hypothesis-driven approach, we extract predefined patterns from learners’ interactions with the course materials. We then identify and analyze different longitudinal profiles among learners by clustering their study pattern sequences during the course. Our second method is a data-driven approach to discover latent study patterns and track them over time in a completely unsupervised manner. We propose a clustering pipeline to model and cluster activity sequences at each time step and then search for matching clusters in previous steps to enable tracking over time. The proposed pipeline is general and allows for analysis at different levels of action granularity and time resolution in various online learning environments. Experiments with synthetic data show that our proposed method can accurately detect latent study patterns and track changes in learning behaviours. We demonstrate the application of both methods on a MOOC dataset and study the temporal dynamics of learners’ behaviour in this context.

Download Full-text

Representational drift in the mouse visual cortex

10.1101/2020.10.05.327049 ◽

2020 ◽

Cited By ~ 1

Author(s):

Daniel Deitch ◽

Alon Rubin ◽

Yaniv Ziv

Keyword(s):

Large Scale ◽

Activity Patterns ◽

Population Level ◽

Population Activity ◽

Cortical Areas ◽

Electrophysiological Recordings ◽

The Face ◽

Visual Cortical ◽

Mouse Visual Cortex ◽

Over Time

AbstractNeuronal representations in the hippocampus and related structures gradually change over time despite no changes in the environment or behavior. The extent to which such ‘representational drift’ occurs in sensory cortical areas and whether the hierarchy of information flow across areas affects neural-code stability have remained elusive. Here, we address these questions by analyzing large-scale optical and electrophysiological recordings from six visual cortical areas in behaving mice that were repeatedly presented with the same natural movies. We found representational drift over timescales spanning minutes to days across multiple visual areas. The drift was driven mostly by changes in individual cells’ activity rates, while their tuning changed to a lesser extent. Despite these changes, the structure of relationships between the population activity patterns remained stable and stereotypic, allowing robust maintenance of information over time. Such population-level organization may underlie stable visual perception in the face of continuous changes in neuronal responses.

Download Full-text

Spatio-temporal dynamics of dengue in Brazil: seasonal travelling waves and determinants of regional synchrony

10.1101/475335 ◽

2018 ◽

Author(s):

Mikhail Churakov ◽

Christian J. Villabona-Arenas ◽

Moritz U.G. Kraemer ◽

Henrik Salje ◽

Simon Cauchemez

Keyword(s):

Large Scale ◽

Temporal Dynamics ◽

Mosquito Control ◽

Human Mobility ◽

Travelling Waves ◽

Seasonal Pattern ◽

Temporal Analysis ◽

Vector Ecology ◽

Spatio Temporal ◽

Dengue Epidemics

AbstractDengue continues to be the most important vector-borne viral disease globally and in Brazil, where more than 1.4 million cases and over 500 deaths were reported in 2016. Mosquito control programmes and other interventions have not stopped the alarming trend of increasingly large epidemics in the past few years.Here, we analyzed monthly dengue cases reported in Brazil between 2001 and 2016 to better characterize the key drivers of dengue epidemics. Spatio-temporal analysis revealed recurring travelling waves of disease occurrence. Using wavelet methods, we characterised the average seasonal pattern of dengue in Brazil, which starts in the western states of Acre and Rondônia, then travels eastward to the coast before reaching the northeast of the country. Only two states in the north of Brazil (Roraima and Amapá) did not follow the countrywide pattern and had inconsistent timing of dengue epidemics throughout the study period.We also explored epidemic synchrony and timing of annual dengue cycles in Brazilian regions. Using gravity style models combined with climate factors, we showed that both human mobility and vector ecology contribute to spatial patterns of dengue occurrence.This study offers a characterization of the spatial dynamics of dengue in Brazil and its drivers, which could inform intervention strategies against dengue and other arboviruses.Author summaryIn this paper we studied the synchronization of dengue epidemics in Brazilian regions. We found that a typical dengue season in Brazil can be described as a wave travelling from the western part of the country towards the east, with the exception of the two most northern equatorial states that experienced inconsistent seasonality of dengue epidemics.We found that the spatial structure of dengue cases is driven by both climate and human mobility patterns. In particular, precipitation was the most important factor for the seasonality of dengue at finer spatial resolutions.Our findings increase our understanding of large scale dengue patterns and could be used to enhance national control programs against dengue and other arboviruses.

Download Full-text

Visual Literacy and Virtual Rephotography

Visual Literacy in The Virtual Realm: The Book of Selected Readings 2021 ◽

10.52917/ivlatbsr.2021.018 ◽

2021 ◽

pp. 81-93

Author(s):

Gary McLeod

Keyword(s):

Online Learning ◽

Video Game ◽

Visual Literacy ◽

Large Scale ◽

Google Street View ◽

Street View ◽

Northeastern Japan ◽

Vantage Points ◽

Over Time

Rephotography is a varied set of practices that begin with taking one or more pictures of the same subject. Valued for generating conversations in-situ about a place over time, recent large-scale migration to online learning draws attention to rephotography’s virtual modes. From examples that use online location software (e.g., Google Street View) to those in video game worlds, virtual rephotography might present convenient windows to unreachable destinations. However, rephotographing without having visited actual vantage points needs to take into account complexity and disjointedness introduced by such tools. Drawing from the author’s current practice-led research into photomedia, visual literacy and temporality in Northeastern Japan, emergent particularities are discussed for developing visual literacy through a necessary application of Google Street View.

Download Full-text

Trajectories of engagement: a repeated cross-sectional investigation of student perceptions of an online learning environment

Research in Learning Technology ◽

10.3402/rlt.v20i0.17143 ◽

2012 ◽

Vol 20 ◽

Cited By ~ 5

Author(s):

Stuart Palmer ◽

Dale Holt

Keyword(s):

Online Learning ◽

Learning Environment ◽

Student Perceptions ◽

Large Scale ◽

Online Learning Environment ◽

Cross Sectional ◽

Short Period ◽

Significant Difference ◽

System Use ◽

Over Time

Evaluations of online learning environments (OLEs) often present a snapshot of system use. It has been identified in the literature that extended evaluation is required to reveal statistically significant developments in the evolution of system use over time. The research presented here draws on student OLE evaluations surveys run over the period 20042011 and include nearly 6800 responses exploring students’ perceptions of importance of, and satisfaction with elements of their OLE. Across the survey period, satisfaction ratings with all OLE elements rose significantly, suggesting a positive student engagement with the OLE over time. The corresponding ratings of importance of OLE elements generally rose significantly, though a number of elements registered no significant difference in the first two years of the survey, suggesting that short period surveys may struggle to reveal statistically significant trends. OLE element use appeared to be closely linked to perceived value. The OLE elements with the highest mean importance and satisfaction ratings related to student access of online learning resources. Other detailed results are also reported. We demonstrate a method for, and one large-scale case study of, quantifying and visualising the trajectories of engagement that students have had with an institutional OLE over time.Keywords: online learning environment; learning management system; repeated cross-sectional evaluation; student survey(Published: 24 September 2012)Citation: Research in Learning Technology 2012, 20: 17143 - http://dx.doi.org/10.3402/rlt.v20i0.17143

Download Full-text

Capturing the spatiotemporal dynamics of task-initiated thoughts with combined EEG and fMRI

10.1101/346346 ◽

2018 ◽

Cited By ~ 3

Author(s):

Lucie Bréchet ◽

Denis Brunet ◽

Gwénaël Birot ◽

Rolf Gruetter ◽

Christoph M. Michel ◽

...

Keyword(s):

Large Scale ◽

Temporal Dynamics ◽

Mental Arithmetic ◽

Brain Activity ◽

Activity Patterns ◽

Mental States ◽

Spatiotemporal Dynamics ◽

Source Analysis ◽

Eeg Source Analysis ◽

Large Scale Networks

AbstractWhen at rest, our mind wanders from thought to thought in distinct mental states. Despite the marked importance of ongoing mental processes, it is challenging to capture and relate these states to specific cognitive contents. In this work, we employed ultra-high field functional magnetic resonance imaging (fMRI) and high-density electroencephalography (EEG) to study the ongoing thoughts of participants instructed to retrieve self-relevant past episodes for periods of 20s. These task-initiated, participant-driven activity patterns were compared to a distinct condition where participants performed serial mental arithmetic operations, thereby shifting from self-related to self-unrelated thoughts. BOLD activity mapping revealed selective activity changes in temporal, parietal and occipital areas (“posterior hot zone”), evincing their role in integrating the re-experienced past events into conscious representations during memory retrieval. Functional connectivity analysis showed that these regions were organized in two major subparts of the default mode network, previously associated to “scene-reconstruction” and “self-experience” subsystems. EEG microstate analysis allowed studying these participant-driven thoughts in the millisecond range by determining the temporal dynamics of brief periods of stable scalp potential fields. This analysis revealed selective modulation of occurrence and duration of specific microstates in both conditions. EEG source analysis revealed similar spatial distributions between the sources of these microstates and the regions identified with fMRI. These findings support growing evidence that specific fMRI networks can be captured with EEG as repeatedly occurring, integrated brief periods of synchronized neuronal activity, lasting only fractions of seconds.SignificanceWe investigated the spatiotemporal dynamics of large-scale brain networks related to specific conscious thoughts. We demonstrate here that instructing participants to direct their thoughts to either episodic autobiographic memory or to mental arithmetic modulates distinct networks both in terms of highly spatially-specific BOLD signal oscillations as well as fast sub-second dynamics of EEG microstates. The combined findings from the two modalities evince a clear link between hemodynamic and electrophysiological signatures of spontaneous brain activity by the occurrence of thoughts that last for fractions of seconds, repeatedly appearing over time as integrated coherent activities of specific large-scale networks.

Download Full-text

Multiple concurrent feedforward and feedback streams in a cortical hierarchy

10.1101/2021.01.04.425277 ◽

2021 ◽

Author(s):

Elham Barzegaran ◽

Gijs Plomp

Keyword(s):

Large Scale ◽

Temporal Dynamics ◽

Activity Patterns ◽

Contrast Effects ◽

Hierarchical Processing ◽

Cortical Areas ◽

Cortical Hierarchy ◽

Functional Hierarchy ◽

Feedback Networks ◽

Insight Into

AbstractVisual stimuli evoke fast-evolving activity patterns that are distributed across multiple cortical areas. These areas are hierarchically structured, as indicated by their anatomical projections, but how large-scale feedforward and feedback streams are functionally organized in this system remains an important missing clue to understanding cortical processing. By analyzing visual evoked responses in laminar recordings from six cortical areas in awake mice, we established the simultaneous presence of two feedforward and two feedback networks, each with a distinct laminar functional connectivity profile, frequency spectrum, temporal dynamics and functional hierarchy. We furthermore identified a distinct role for each of these four processing streams, by leveraging stimulus contrast effects and analyzing receptive field convergency along functional interactions. Our results support a dynamic dual counterstream view of hierarchical processing and provide new insight into how separate functional streams can simultaneously and dynamically operate in visual cortex.

Download Full-text

Word embeddings quantify 100 years of gender and ethnic stereotypes

Proceedings of the National Academy of Sciences ◽

10.1073/pnas.1720347115 ◽

2018 ◽

Vol 115 (16) ◽

pp. E3635-E3644 ◽

Cited By ~ 85

Author(s):

Nikhil Garg ◽

Londa Schiebinger ◽

Dan Jurafsky ◽

James Zou

Keyword(s):

United States ◽

Machine Learning ◽

Temporal Dynamics ◽

Temporal Analysis ◽

The United States ◽

English Word ◽

Attitudes Toward Women ◽

Word Embeddings ◽

Asian Immigration ◽

Over Time

Word embeddings are a powerful machine-learning framework that represents each English word by a vector. The geometric relationship between these vectors captures meaningful semantic relationships between the corresponding words. In this paper, we develop a framework to demonstrate how the temporal dynamics of the embedding helps to quantify changes in stereotypes and attitudes toward women and ethnic minorities in the 20th and 21st centuries in the United States. We integrate word embeddings trained on 100 y of text data with the US Census to show that changes in the embedding track closely with demographic and occupation shifts over time. The embedding captures societal shifts—e.g., the women’s movement in the 1960s and Asian immigration into the United States—and also illuminates how specific adjectives and occupations became more closely associated with certain populations over time. Our framework for temporal analysis of word embedding opens up a fruitful intersection between machine learning and quantitative social science.

Download Full-text

Multiplexed measurement of protein biomarkers in high-frequency longitudinal dried blood spot (DBS) samples: characterization of inflammatory responses

10.1101/643239 ◽

2019 ◽

Cited By ~ 2

Author(s):

N. Leigh Anderson ◽

Morteza Razavi ◽

Matthew E. Pope ◽

Richard Yip ◽

Terry W. Pearson

Keyword(s):

Temporal Dynamics ◽

Inflammatory Responses ◽

Temporal Analysis ◽

Dried Blood Spots ◽

Blood Protein ◽

Spectrometric Method ◽

Mass Spectrometric Method ◽

Learning Approaches ◽

Protein Biomarkers ◽

Over Time

AbstractA detailed understanding of changes in blood protein biomarkers occuring in individuals over time would enable truly personalized approaches to health and disease monitoring. Such measurements could reveal smaller, earlier departures from normal baseline levels of biomarkers thus allowing better disease detection and treatment monitoring. Current practice, however, generally involves infrequent, sporadic biomarker testing, and this undersampling likely fails to capture important biological phenomena. Here we report the use of a robust multiplex immuno-mass spectrometric method (SISCAPA) to measure a panel of clinically-relevant proteins in a unique collection of 1,522 dried blood spots collected longitudinally by 8 individuals over periods of up to 9 years, with daily sampling during some intervals. Analytical workflow CVs of 2-6% for most assays were achieved by normalizing DBS plasma volume using a set of 3 minimally varying proteins, facilitating temporal analysis of both high- and low-amplitude biomarker changes compared to personalized baselines. The biomarkers included a panel of 9 positive and 5 negative acute phase response (inflammatory) proteins, allowing longitudinal analysis of inflammation markers associated with major and minor infections, influenza vaccination, recovery from hip-replacement surgery and Crohn’s disease. The results illustrate complex time-dependent “biomarker trajectories” on multiple timescales and provide a basis for detailed personalized models of inflammation dynamics. The striking stability of most biomarker protein levels over time, combined with the convenience of self-sampling and low cost of multiplexed measurements using mass spectrometry, provide a new window into the temporal dynamics of disease processes. The extensive results obtained using this high throughput approach offer a new source of precision biomarker ‘big data’ amenable to machine learning approaches and application to more personalized health monitoring.

Download Full-text

Intelligent Drought Tracking for its Use in Machine Learning: Implementation and First Results

10.29007/klgg ◽

2018 ◽

Cited By ~ 1

Author(s):

Vitali Diaz ◽

Gerald A. Corzo Perez ◽

Henny A.J. Van Lanen ◽

Dimitri Solomatine

Keyword(s):

Machine Learning ◽

Temporal Dynamics ◽

Temporal Analysis ◽

Drought Monitoring ◽

Grid Data ◽

Drought Prediction ◽

Documented Information ◽

First Results ◽

Spatio Temporal ◽

Over Time

Due to the underlying characteristics of drought, monitoring of its spatio-temporal development is difficult. Last decades, drought monitoring have been increasingly developed, however, including its spatio-temporal dynamics is still a challenge. This study proposes a method to monitor drought by tracking its spatial extent. A methodology to build drought trajectories is introduced, which is put in the framework of machine learning (ML) for drought prediction. Steps for trajectories calculation are (1) spatial areas computation, (2) centroids localization, and (3) centroids linkage. The spatio- temporal analysis performed here follows the Contiguous Drought Area (CDA) analysis. The methodology is illustrated using grid data from the Standardized Precipitation Evaporation Index (SPEI) Global Drought Monitor over India (1901-2013), as an example. Results show regions where drought with considerable coverage tend to occur, and suggest possible concurrent routes. Tracks of six of the most severe reported droughts were analysed. In all of them, areas overlap considerably over time, which suggest that drought remains in the same region for a period of time. Years with the largest drought areas were 2000 and 2002, which coincide with documented information presented. Further research is under development to setup the ML model to predict the track of drought.

Download Full-text

Dynamics of neural microstates in the VTA-striatal-prefrontal loop during novelty exploration in the rat

10.1101/2020.08.27.270249 ◽

2020 ◽

Author(s):

A. Mishra ◽

N. Marzban ◽

M. X Cohen ◽

B. Englitz

Keyword(s):

Spatial Patterns ◽

Neural Activity ◽

Large Scale ◽

Temporal Dynamics ◽

Activity Patterns ◽

Field Potential ◽

Brain Regions ◽

Brain Areas ◽

Neural Recordings ◽

Temporal Correlations

AbstractEEG microstates refer to quasi-stable spatial patterns of scalp potentials, and their dynamics have been linked to cognitive and behavioral states. Neural activity at single and multiunit levels also exhibit spatiotemporal coordination, but this spatial scale is difficult to relate to EEG. Here, we translated EEG microstate analysis to triple-area local field potential (LFP) recordings from up to 192 electrodes in rats to investigate the mesoscopic dynamics of neural microstates within and across brain regions.We performed simultaneous recordings from the prefrontal cortex (PFC), striatum (STR), and ventral tegmental area (VTA) during awake behavior (object novelty and exploration). We found that the LFP data can be accounted for by multiple, recurring, quasi-stable spatial activity patterns with an average period of stability of ~60-100 ms. The top four maps accounted for 60-80% of the total variance, compared to ~25% for shuffled data. Cross-correlation of the microstate time-series across brain regions revealed rhythmic patterns of microstate activations, which we interpret as a novel indicator of inter-regional, mesoscale synchronization. Furthermore, microstate features, and patterns of temporal correlations across microstates, were modulated by behavioural states such as movement and novel object exploration. These results support the existence of a functional mesoscopic organization across multiple brain areas, and open up the opportunity to investigate their relation to EEG microstates, of particular interest to the human research community.Significance StatementThe coordination of neural activity across the entire brain has remained elusive. Here we combine large-scale neural recordings at fine spatial resolution with the analysis of microstates, i.e. short-lived, recurring spatial patterns of neural activity. We demonstrate that the local activity in different brain areas can be accounted for by only a few microstates per region. These microstates exhibited temporal dynamics that were correlated across regions in rhythmic patterns. We demonstrate that these microstates are linked to behavior and exhibit different properties in the frequency domain during different behavioural states. In summary, LFP microstates provide an insightful approach to studying both mesoscopic and large-scale brain activation within and across regions.

Download Full-text