Detailseite
Projekt Druckansicht

Augenbewegungen beim gedankenverlorenen Lesen: Experimente und computationale Modellierung

Fachliche Zuordnung Allgemeine, Kognitive und Mathematische Psychologie
Förderung Förderung von 2007 bis 2012
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 43741491
 
Erstellungsjahr 2012

Zusammenfassung der Projektergebnisse

It sometimes happens that we finish reading a page of text just to realize that we have no idea what we just read. During these episodes of mindless reading our mind is elsewhere yet the eyes still move across the text. Recent progress in understanding this elusive phenomenon has been achieved by using subjective self-report measures that have treated mindless reading as an all-or-none phenomenon. In the current project, we investigated the full range of different levels of mindless reading in an eye tracking laboratory based on several experimental approaches: (a) z-string scanning, (b) reading lists of shuffled words, (c) reading with a distractor task, (d) skimming of text, (e) reading text in an unknown language, and (f) reading very long passages of boring texts during sessions of several hours. In these paradigms, we extended previous research aiming at approximating states of mind wandering via experimental manipulations that reduce cognitive text processing by using diverse approaches to manipulate higher-level text processing during reading-like tasks (a-e). Moreover, we developed an objective behavioral measure to catch episodes when mindless reading spontaneously occur during a normal reading task (f): the sustained attention to stimulus task (SAST) measures mindless reading via overlooking of errors in the text, and via low sensitivity in signal detection analyses of error detection. Using advanced statistical analyses (linear mixed effects models), we observed strong effects of mindlessness visible in measures of eye-movement control. As a consistent finding, influences from word processing and from linguistic processing on eye movements were reduced during mindless reading across paradigms. Based on these effects, we demonstrated in a Bayesian analysis that it is possible to predict states of mindless reading from eye movement recordings, which may trigger the development of new methods to measure mindless reading based on objective online-recordings. To explain findings on reduced cognitive processing, we introduced the levels-of-inattention hypothesis that mindless reading, and mind wandering in general, is a graded process that occurs at different levels of cognitive processing, and reflects episodes of early versus late attentional selection. We found support for the levels-of-inattention hypothesis in different paradigms based on analyses of eye movements and the overlooking of different kinds of errors in the text. Additionally, we performed numerical simulations of a computational model of eyemovement control on two of the experimental paradigms, and found that mindless reading reduces specific interactions between word processing and saccadic control. These findings imply strong visuomotor control of eye movements in the absence of higher-level processing, and provide support to the concept of an activation field guiding eye movements during reading and non-reading tasks. Moreover, we observed how cognitive-saccadic interactions depended on visual attention allocation in one of our experimental paradigms (b), and used an advanced version of a computational model incorporating the zoom lens of attention to simulate these findings. The simulation results supported our qualitative predictions on an involvement of an adaptive attentional zoom lens and demonstrated that the zoom lens explains many important reading phenomena. Interestingly, we provided the first mathematical model of skipping benefits, an effect that could not be explained by previous eye movement models. Overall, our analyses demonstrated that mindless reading – an elusive phenomenon that has long escaped scientific investigation – can be traced down in the cognitive laboratory and be studied and understood using tools (like eye tracking and statistical and mathematical modeling) and theoretical concepts (like levels of processing and saliency-maps) of cognitive science.

Projektbezogene Publikationen (Auswahl)

  • (2009). Mindless reading revisited: An analysis based on the SWIFT model of eye-movement control. Vision Research, 49, 322-336
    Nuthmann, A., & Engbert, R.
    (Siehe online unter https://doi.org/10.1016/j.visres.2008.10.022)
  • (2010) Eye movements during reading of randomly shuffled texts. Vision Research, 50, 2600-2616
    Schad, D. J., Nuthmann, A., & Engbert, R.
    (Siehe online unter https://doi.org/10.1016/j.visres.2010.08.005)
  • (2012). The zoom lens of attention: Simulating shuffled versus normal text reading using the SWIFT model. Visual Cognition, 20(4-5), 391-421
    Schad, D. J., & Engbert, R.
    (Siehe online unter https://doi.org/10.1080/13506285.2012.670143)
  • (2012). Your mind wanders weakly, your mind wanders deeply: Objective measures reveal mindless reading at different levels. Cognition (August 2, 2012)
    Schad, D. J., Nuthmann, A., & Engbert, R.
    (Siehe online unter https://doi.org/10.1016/j.cognition.2012.07.004)
 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung