A neural model of saccadic eye movement control explains task-specific adaptation

Abstract

Multiple brain learning sites are needed to calibrate the accuracy of saccadic eye movements. This is true because saccades can be made reactively to visual cues, attentively to visual or auditory cues, or planned in response to memory cues using visual, parietal, and prefrontal cortex, as well as superior colliculus, cerebellum, and reticular formation. The organization of these sites can be probed by displacing a visual target during a saccade. The resulting adaptation typically shows incomplete and asymmetric transfer between different tasks. A neural model of saccadic system learning is developed to explain these data, as well as data about saccadic coordinate changes.

Neural dynamics of saccadic and smooth pursuit eye movement coordination during visual tracking of unpredictably moving targets

“…It unifiesd and further developed the SAC model Grossberg et al, (1997) and Gancarz and Grossberg (1999), and the SPEM model of Pack et al, (2001). Table 1 summarizes the neuroanatomical connections and their functional interpretation, discussed below, that are unified within this model.…”

“…Existing models of saccadic eye movement (Brown, Bullock, & Grossberg, 2004;Dominey & Arbib, 1992;Droulez & Berthoz, 1991;Gancarz & Grossberg, 1999;Grossberg & Kuperstein, 1986;Grossberg, Roberts, Aguilar, & Bullock, 1997;Optican & Quaia, 2002;Waitzman, Ma, Optican, & Wurtz, 1991) exclusively treat the control of saccades. The Grossberg et al (1997) model predicts how a multi-modal map is learned in superior colliculus (SC), such that visual, auditory and planned representations of target positions become aligned in retinal coordinates.…”

“…This replaces the classical notion of a displacement integrator (Jurgens, Becker, & Kornhuber, 1981), which uses the desired change in eye position instead of the desired eye position as an input to the saccade generator. A major difference between the Optican and Quaia (2002) and Gancarz and Grossberg (1999) models lies in how the CBM corrects for saccadic errors. The Optican and Quaia model converts a feedback signal about saccade velocity into a retinotopically organized map in vermis.…”

“…By manipulating the size of the step and/or the speed of the ramp motion, the paradigm induces a monkey to initiate a catch-up saccade or to begin pursuit without an initial saccade. Notably, catch-up saccades depend on both the target's step size and ramp speed thereafter (de Brouwer, Missal, Barnes, & Lefevre, 2002;de Brouwer, Missal, & Lefevre, 2001;de Brouwer, Yuksel, Blohm, Missal, & Lefevre, 2002).Existing models of saccadic eye movement (Brown, Bullock, & Grossberg, 2004;Dominey & Arbib, 1992;Droulez & Berthoz, 1991;Gancarz & Grossberg, 1999;Grossberg & Kuperstein, 1986;Grossberg, Roberts, Aguilar, & Bullock, 1997;Optican & Quaia, 2002;Waitzman, Ma, Optican, & Wurtz, 1991) exclusively treat the control of saccades. The Grossberg et al (1997) model predicts how a multi-modal map is learned in superior colliculus (SC), such that visual, auditory and planned representations of target positions become aligned in retinal coordinates.…”

“…It and Brown et al (2004) also treat how voluntary gating by substantia nigra determines whether excitation of an SC map locus leads to genesis of a corresponding saccade. Gancarz and Grossberg (1999) proposed how several brain regions work together to control and adapt saccades in response to different conditions, notably how the cerebellum (CBM) adapts saccadic gain in a context-sensitive way. Optican and Quaia (2002) proposed that the SC encodes the desired sensory consequence of the saccade in retinotopic coordinates.…”