Abstract
In saccade sequences endpoint errors pose a problem for subsequent saccades in the absence of visual feedback. Compensation for endpoint variability has previously been demonstrated in double step saccades (DSS; e.g. Joiner et al., 2010) and is thought to rely on a copy of the saccade motor vector (corollary discharge; CD). However, these studies typically use highly repetitive stimuli (e.g. one or few identical target vectors). While this eases analysis, it calls into question the generalizability of the findings due to the high target predictability. We present a new, random walk based DSS paradigm (random target vector amplitudes and directions) and a direct way of analyzing this data to provide a more complete, realistic and generalizable description of error compensation in saccade sequences. We regressed the difference vector between the endpoint of the second saccade and the endpoint of a second saccade that does not take first saccade error into account on the vector necessary for full compensation. This provides a direct and complete quantitative estimation of error compensation in DSS's. Data modeling verified the validity of the paradigm and data analysis method. As expected we observed error compensation lower than but comparable to traditional experiments. We also employed this paradigm to replicate previous findings that showed compensation for systematic undershoots after specific-vector saccade adaptation. This indicates that the CD signal used for estimating post-saccadic target location is taken downstream from the site of adaptation (Tanaka, 2003). Utilizing the random walk paradigm for saccade adaptation by Rolfs et al. (2010) together with our random walk DSS paradigm we now also demonstrated this for global saccade adaptation. We developed a new, generalizable DSS paradigm with and successfully employed it to verify, replicate and extend previous findings, demonstrating that endpoint errors are compensated for even in variable stimulus displays.
Meeting abstract presented at VSS 2016