Figure 2.

Experimental design and results from Experiment 1 on feedforward adaptation. a, Schedule of perturbations in Experiment 1. Conditions of interest are the gradual and abrupt perturbations. b, d, Behavioral results from noncatch trials for gradual and abrupt perturbations, respectively. Productions are grouped into bins of 10. Error bars indicate SE. Light gray represents probes (baseline, early, late, washout). The perturbation schedule is shown below the behavioral results (black). There was no difference in results between the two conditions. f, Control (red) and patient (blue) productions in the early probe, late probe, and washout phases, normalized to the baseline values. *Significant difference from baseline. c, e, Productions during the catch trials in the gradual and abrupt conditions, respectively. In these trials, loud speech-shaped noise was used to mask feedback from the participants' speech. On these trials, participants responded by dramatically increasing F1. This type of increase is typical in loud noise and makes it problematic to measure an adaptive response. Nonetheless, the trend in both cases is that CD patients produce a reduced increase in F1 compared with healthy controls.