350 likes | 573 Views
Toward Fully Automated Person-Independent Detection of Mind Wandering. Robert Bixler & Sidney D’Mello rbixler@nd.edu University of Notre Dame July 10, 2013. mind wandering. indicates waning attention occurs frequently 20-40% of the time decreases performance comprehension memory.
E N D
Toward Fully Automated Person-Independent Detection of Mind Wandering Robert Bixler & Sidney D’Mello rbixler@nd.edu University of Notre Dame July 10, 2013
mind wandering • indicates waning attention • occurs frequently • 20-40% of the time • decreases performance • comprehension • memory
solutions • proactive • mindfulness training • Mrazek (2013) • tailoring learning environment • Kopp, Bixler, D’Mello (2014) • reactive • mind wandering detection
related work – attention • Attention and Selection in Online Choice Tasks • Navalpakkam et al. (2012) • Multi-mode Saliency Dynamics Model for Analyzing Gaze and Attention • Yonetani, Kawashima, and Matsuyama (2012) • distinct from mind wandering
mind wandering detection • neural activity • physiology • acoustic/prosodic • eye movements
neural activity Experience Sampling During fMRI Reveals Default Network and Executive System Contributions to Mind Wandering • Christoff et al. (2009)
physiology Automated Physiological-Based Detection of Mind Wandering during Learning • Blanchard, Bixler, D’Mello(2014)
acoustic-prosodic In the Zone: Towards Detecting Student Zoning Out Using Supervised Machine Learning • Drummond and Litman (2010)
eye movements mindless reading mindful reading
research questions • can mind wandering be detected from eye gaze data? • which features are most useful for detecting mind wandering?
data collection • 4 texts on research methods • self-paced page-by-page • 30-40 minutes • difficulty and value • auditory probes • 9 per text • inserted psuedorandomly (4-12s) tobii tx300
data analysis • compute fixations • OGAMA (Open Gaze and Mouse Analyzer) (Voßkühler et al. 2008) • compute features • build supervised machine learning models
features • global • local • context
global features • eye movements • fixation duration • saccade duration • saccade length • fixation dispersion • reading depth • fixation/saccade ratio
local features • reading patterns • word length • hypernym depth • number of synonyms • frequency • fixation type • regression • first pass • single • gaze • no word
context features • positional timing • since session start • since text start • since page start • previous page times • average • previous page to average ratio • task • difficulty • value
supervised machine learning • parameters • window size (4, 8, or 12) • minimum number of fixations (5, 1/s, 2/s, or 3/s) • outlier treatment (trimmed, winsorized, none) • feature type (global, local, context, combined) • downsampling • feature selection • classifiers (20 standard from weka) • leave-several-subjects-out cross validation (66:34 split)
1. can mind wandering be detected using eye gaze data? confusion matrices end-of-page within-page
2. which features are most useful for detecting mind wandering?
2. which features are most useful for detecting mind wandering?
summary • mind wandering detection is possible • kappas of .28 to .17 • end-of-page models performed better • global features were best • exception: context features highest ranked for end-of-page
enhanced feature set • global • pupil diameter • blink frequency • saccade angle • local • cross-line saccades • end-of-clause fixations
what does mind wandering look like? • saccades • slower • shorter • more frequent blinks • larger pupil diameters
limitations • eye tracker cost • population validity • self-report • classification accuracy
future work • multiple modalities • different types of mind wandering • mind wandering intervention
acknowledgements • Blair Lehman • Art Graesser • Jennifer Neale • Nigel Bosch • Caitlin Mills