Attention shift and remapping across saccades
by Tao Yao
Date of Examination:2016-12-19
Date of issue:2017-03-15
Advisor:Prof. Dr. Stefan Treue
Referee:Prof. Dr. Stefan Treue
Referee:Prof. Dr. Uwe Mattler
Referee:Prof. Dr. Ralf Heinrich
Referee:Prof. Dr. Tim Gollisch
Referee:Prof. Dr. Alexander Gail
Referee:Prof. Dr. Andreas Stumpner
Files in this item
Name:Yao PhD thesis.pdf
Size:11.3Mb
Format:PDF
Description:PhD dissertation
Abstract
English
Humans and monkeys make two to three saccades per second on average when they are awake, and they are able to keep track of relevant visual stimuli while making saccadic eye-movements to scan a visual scene. Since the visual system mostly operates using retinotopic representations, a rapid, saccade-synchronized shift of top-down attentional modulation from the pre-saccadic to the post-saccadic target population of neurons is optimal. The time-course of this shift of top-down attention across saccades has never been explicitly measured. In the first study, we recorded the extracellular activity of single neurons in area MT of two macaque monkeys during a task that required them to make a saccade while maintaining attention on one of four moving random dot patterns (RDPs). We investigated how attention modulation emerged and decayed in the neuronal activity in area MT across saccades. We found, for the first time, that trans-saccadic attentional shift is well-synchronized to the saccade. In the second study, by using a similar paradigm with finer temporal resolution, we also showed that the top-down spatial attention is fully available at the task-relevant location within 30 milliseconds after the saccade in human subjects. Further, our data indicated that under our task conditions, subjects very rarely confused a distractor stimulus for the target. Spatial attention and saccadic execution thus appeared to co-ordinate well to ensure that relevant objects were attentionally enhanced soon after the beginning of each eye fixation. Additionally, we did not find any evidence for a predictive, pre-saccadic shift of attention to the post-saccadic target population when the human subjects were doing the task. This is consistent with our first study and others have not found evidence for “predictive” pre-saccadic attentional remapping to the post-saccadic target population. In the third study, we investigated the remapping response in MT. To do this, we designed a new remapping paradigm and recorded neuronal activity. The monkeys in this study were required to perform a task very similar to the first study. In a very important condition, the stimuli were removed just before the saccade onset so that it never appeared in the neuron’s RF. We found evidence for a memory trace in MT. The memory trace was modulated by top-down spatial attention. This was the first demonstration of the top-down attention influence on remapping activity. We found only a small and transient effect of motion direction on the memory trace in only one monkey, suggesting that the remapped response did not signal the motion direction of the stimulus in the RF before the saccade. Our results support the hypothesis that the memory trace emerges from the rapid shift of attention to the post-saccadic neuronal population and the resultant enhancement of the baseline firing-rate. In summary, our physiological data combined with our human psychophysics results support our hypothesis that spatial attention and saccadic processing co-ordinate well to ensure that relevant locations are attentionally enhanced soon after the beginning of each eye fixation, and can be tracked and rapidly processed across saccades.
Keywords: attention; saccades; eye movements; monkey; electrophysiology; psychophysics; remapping; human; vision