Integration and Segregation for Perception and Smooth Pursuit

Smooth pursuit eye movements are used to follow moving objects. Traditionally, smooth pursuit research has studied how the brain generates eye movements to follow small visual targets that move in isolation. However, objects in a natural scene often have components that move in different directions or at different speeds such as wings, legs, arms and tails. To  generate a single velocity command for smooth pursuit, motion information from all components of the object must be integrated.   

Our laboratory studies how signals arising from retinal stimulation with moving, spatially extended objects are integrated across multiple motion detectors and transformed into a single eye velocity command. One type of stimulus that we use is the random dot cinematogram (RDC), which is used to study visual motion processing. The direction and speed of individual component dots in an RDC can be varied independently, allowing us to manipulate the type of stimulation that each unit in a population of motion detectors receives. Smooth pursuit to RDCs inmproves pursuit and reduces saccades during pursuit, and is driven by the global percept of the stimulus and not the individual dot elements (7). A movie of an RDC in which all dots move in the same direction and with the same speed is shown below.



A goal of our work is to determine how signals arising in multiple motion detectors in the perceptual system are transformed to a single eye velocity command for smooth pursuit. A technique to assess this is to add noise to an RDC, and compare the effect of the noise on the motion perception system with its effect on the smooth pursuit system. We find that the response of both systems to stimuli with added noise is remarkably similar, indicating that while the noise in the perceptual system is magnified when pursuit is engaged, no new noise sources are added by the pursuit system (8).

                                              Perception and pursuit show similar breakpoints when noise is added to the stimulus


Natural objects that we pursue often stimulate peripheral retina because they are larger than the standard spot pursuit stimulus.  Furthermore, natural objects also have features.  Because large RDC stimuli impove pursuit and reduce saccades, we wondered if a benefit provided by the peripheral motion that natural objects afford might be to facilitate the inspection of features.  To test this, we devised a pursuit task stimulus in which the "features" were four dots located around a central pursuit target in a "+" configuration.  Observers were required to identify one of the dots when it dimmed briefly during the trial.

Our pursuit feature identification task with and without an RDC pursuit stimulus



We initially had observers pursue the stimulus alone and found that introducing a "gap" by removing the central spot for 200 msec before the target spot dimmed resulted in better task performance.  We reasoned that attention was being devoted to the spot during pursuit, and that removing it freed up attention for the task.  Then we asked if using the background as an alternative drive to the pursuit system might also free up attention since it improves eye movements and reduces saccades, possibly rendering use of the spot unnecssary.  This was the result, suggesting that attention could be released from the spot even when it remained physically present.
Fewer saccades and better "feature identifcation" with RDC pursuit stimuli


Eye movements while pursuing and performing the identification task with and without the RDC background. A)15 representative horizontal (left) and vertical (right) eye velocity traces for one observer (HH) randomly sampled from two blocks of trials, one in which the RDC background was on and another in which it was off. Target speed was 30 deg/sec. Top traces (red) are from background-off trials; bottom traces (blue) are from background-on trials. Note fewer saccadic intrusions with the background on. Saccade traces were truncated at +/- 50 deg/sec for ease of viewing. B)Number of saccadic intrusions per trial for each observer averaged over all background-on/off trial blocks. C)Task performance during pursuit with the RDC background on and off. Identification accuracy was better for all observers and at most target speeds and dimming times when the background was present.


Home | Vision for Action | Movement Control | Timing | Integration & Segregation |

| People | Life at SKERI | Related Links | Publications | Contact | SK Home |