Figure 6. Evidence that the expression of a first CR is necessary for expression of subsequent second CR.

(A–B) Eyelid CRs on two example trials from ipsilateral and contralateral sequence training sessions respectively. Color-coding of time intervals is preserved from Figures 2 and 3. (C) Scatter plot representing the amplitude of the second CR versus the amplitude of the first CR in ipsilateral sequence training. Each dot represents a single CS-alone trial. Vertical and horizontal solid black lines represent non-CRs cutoffs, dotted black line shows a diagonal where second CR amplitude equals that of the first CR. Dots corresponding to example trials from panels (A–B) are shown in orange. (D) Similar to (C), but for subjects trained in contralateral sequence protocol. Data obtained from sessions with different gap intervals are color-coded as indicated in legend. (E) Average probability of the second CR as a function of first CR amplitude. Error-bars show 95% confidence intervals (obtained by bootstrapping with 0.3 mm nonCR threshold, 2000 repetitions). (F) Same as (E), but for subjects trained at contralateral sequence of CRs. Here we combined data from sessions with different gap intervals.