Frontiers in Human Neuroscience (Jul 2016)

Adaptation of saccadic sequences with and without remapping

  • Delphine Lévy-Bencheton,
  • Aarlenne Zein Khan,
  • Denis Pélisson,
  • Caroline Tilikete,
  • Laure Pisella

DOI
https://doi.org/10.3389/fnhum.2016.00359
Journal volume & issue
Vol. 10

Abstract

Read online

It is relatively easy to adapt visually-guided saccades because the visual vector and the saccade vector match. The retinal error at the saccade landing position is compared to the prediction error, based on target location and efference copy. If these errors do not match, planning processes at the level(s) of the visual and/or motor vectors processing are assumed to be inaccurate and the saccadic response is adjusted. In the case of a sequence of two saccades, the final error can be attributed to the last saccade vector or to the entire saccadic displacement. Here, we asked whether and how adaptation can occur in the case of remapped saccades, such as during the classic double-step saccade paradigm, where the visual and motor vectors of the second saccade do not coincide and so the attribution of error is ambiguous. Participants performed saccades sequences to two targets briefly presented prior to first saccade onset. The second saccade target was either briefly re-illuminated (visually-guided paradigm) or not (remapping paradigm) upon first saccade offset. To drive adaptation, the second target was presented at a displaced location (backward or forward jump condition or control – no jump) at the end of the second saccade. Pre- and post-adaptation trials were identical, without the re-appearance of the target after the second saccade. For the 1st saccade endpoints, there was no change as a function of adaptation. For the 2nd saccade, there was a similar increase in gain in the forward jump condition (52% and 61% of target jump) in the two paradigms, whereas the gain decrease in the backward condition was much smaller for the remapping paradigm than for the visually-guided paradigm (41% vs. 94%). In other words, the absolute gain change was similar between backward and forward adaptation for remapped saccades.In conclusion, we show that remapped saccades can be adapted, suggesting that the error is attributed to the visuo-motor transformation of the remapped visual vector. The mechanisms by which adaptation takes place for remapped saccades may be similar to those of forward visually-guided saccades, unlike those involved in adaptation for backward visually-guided saccades.

Keywords