非表示:
キーワード:
-
要旨:
How does our visual system decide where to look? The Linear Approach to Threshold with
Ergodic Rate (LATER: Carpenter, 1995) is a simple decision-making model for saccadic
eye movements. Currently, experimental data suggest that saccadic eye-movements can be
discriminated according to whether they are performed for directed fixations or for item
recognition (Montagnini Chelazzi, 2005; Bieg et al., submitted). Unfortunately, sufficient
goodness-of-fit can only be acquired with large datasets, for each individual participant. Here,
we investigate whether adapting LATER with modern computational methods can allow for
saccades to be classified for their functionality, with minimal data and in real-time. In doing
so, we strive towards the eventual goal of using the LATER model for predicting observer
intentions in real-world applications.
