A Neurocomputational Model of Anticipation and Sustained Inattentional Blindness in Hierarchies

• Anticipation and prediction have been identified as key functions of many brain areas facilitating recognition, perception, and planning. In this chapter we present a hierarchical neurocomputational model in which feedback, effectively predicting or anticipating task-relevant features, leads to sustained inattentional blindness. A psychological experiment on sustained inattentional blindness in human subjects is simulated to provide visual input to a hierarchy of Echo State Networks. Other parts of the model receive input relevant to tracking the attended object and also detecting the unexpected object, feedback from which is then used to simulate engagement in the task and compared to results obtained without feedback, simulating passive observation. We find a significant effect of anticipation enhancing performance at the task and simultaneously degrading detection of unexpected features, thereby modelling the sustained inattentional blindness effect. We therefore suggest that anticipatory /predictive mechanisms are responsible for sustained inattentional blindness.

Ämnesord

NATURVETENSKAP  -- Data- och informationsvetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences (hsv//eng)

Nyckelord

Enaction

Anticipation

Prediction

Neurocomputation

Reservoir Systems

Association

Sustained Inattentional Blindness

Neural Modelling

Cortical Hierarchies

Technology

Teknik

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)