Data from: Cerebellar complex spikes multiplex complementary behavioral information

Fig 1. The influence of repetitive saccade paradigm on saccade behavior and encoding of primary saccade direction by CSs Fig 2. CSs encode primary saccade amplitude and duration but not peak velocity during the ‘early post-saccadic period’ (Note: The values of bootstrapped mean ± confidence intervals in Fig 2B, the right panel, are subject to change with every iteration and may not perfectly match the values of the original figure in the manuscript.) Fig 3. CSs encode corrective saccade direction, amplitude and duration during the ‘post-corrective saccadic period’ (Note: The values of bootstrapped mean ± confidence intervals in Fig 3D, the right panel, are subject to change with every iteration and may not perfectly match the values of the original figure in the manuscript.) Fig 4. CSs encode the retinal error direction and magnitude during the ‘late post-saccadic period’ in a manner different from primary and corrective saccades Fig 5. CS activity during the ‘early post-saccadic period’ carries information on primary saccades and errors, whereas during the ‘late post-saccadic period’ it is mostly error-driven Fig 6. Changes in CS duration encode changes in primary and corrective saccades as well as errors Fig 7. A trial onset related CS discharge seems to predict the upcoming events Fig 8. Encoding of different task parameters by individual PCs SUPPORTING INFORMATION S1 Fig. Influence of the repetitive saccade task on saccade vigor S2 Fig. Relationship of CS response to saccade end, CS responses to CF and CP saccades and a comparison of CF and CP saccades made in the same direction in an exemplary PC (Note: The values of bootstrapped mean ± confidence intervals in S2 Fig C are subject to change with every iteration and may not perfectly match the values of the original figure in the manuscript.) S3 Fig. Analysis of encoding of primary saccade metrics, separately for CSs in PDps and PDps+180° (Note: The values of bootstrapped mean ± confidence intervals in S3 Fig F and H are subject to change with every iteration and may not perfectly match the values of the original figure in the manuscript.) S4 Fig. Analysis of encoding of corrective saccade metrics, separately for CSs in PDcorr and PDcorr+180° (Note: The values of bootstrapped mean ± confidence intervals in S4 Fig G, H, I and J are subject to change with every iteration and may not perfectly match the values of the original figure in the manuscript.) S5 Fig. Primary saccades made in the same direction but resulting in opposite errors evoke similar CS responses S6 Fig. The influence of primary saccade amplitude in PDps and PDps+180° with comparable error sizes S7 Fig. A multiple regression analysis testing the influence of primary saccades and errors on CS activity during non-overlapping post-saccadic periods S8 Fig. Complementary changes in CS duration and firing rate

图1. 重复扫视范式对眼跳行为以及复杂棘波（Complex Spikes, CS）的主眼跳方向编码的影响
图2. 在“眼跳后早期时程”中，CS可编码主眼跳的幅度与持续时间，但无法编码峰值速度（注：图2B右侧面板中的自助法均值±置信区间数值会随每次迭代发生变化，可能与手稿中原始图表的数值不完全一致）
图3. 在“矫正眼跳后时程”中，CS编码矫正眼跳的方向、幅度与持续时间（注：图3D右侧面板中的自助法均值±置信区间数值会随每次迭代发生变化，可能与手稿中原始图表的数值不完全一致）
图4. 在“眼跳后晚期时程”中，CS以区别于主眼跳与矫正眼跳的方式编码视网膜误差的方向与幅度
图5. “眼跳后早期时程”中的CS活动携带主眼跳与误差的相关信息，而“眼跳后晚期时程”中的CS活动则主要由误差驱动
图6. CS持续时间的变化可编码主眼跳、矫正眼跳以及误差的变化
图7. 与试次起始相关的CS放电活动似乎可预测即将发生的事件
图8. 单个浦肯野细胞（Purkinje Cell, PC）对不同任务参数的编码
补充材料
S1图. 重复扫视任务对眼跳活力的影响
S2图. 单示例浦肯野细胞中，CS响应与眼跳结束的关系、CS对同向条件（CF, CP）眼跳的响应，以及同向CF与CP眼跳的比较（注：S2图C中的自助法均值±置信区间数值会随每次迭代发生变化，可能与手稿中原始图表的数值不完全一致）
S3图. 针对主眼跳偏好方向（PDps）与PDps+180°的CS，分别分析其对主眼跳指标的编码情况（注：S3图F与H中的自助法均值±置信区间数值会随每次迭代发生变化，可能与手稿中原始图表的数值不完全一致）
S4图. 针对矫正眼跳偏好方向（PDcorr）与PDcorr+180°的CS，分别分析其对矫正眼跳指标的编码情况（注：S4图G、H、I与J中的自助法均值±置信区间数值会随每次迭代发生变化，可能与手稿中原始图表的数值不完全一致）
S5图. 方向相同但引发相反视网膜误差的主眼跳可诱发相似的CS响应
S6图. 误差大小匹配条件下，主眼跳幅度对主眼跳偏好方向（PDps）与PDps+180°的影响
S7图. 多元回归分析，检验非重叠眼跳后时程中主眼跳与误差对CS活动的影响
S8图. CS持续时间与放电频率的互补变化