Behavioral performance during perceptual-choice trials.

<p><b>(A)</b> Mean probability of T1 choice across all subjects for right and left motion for different values of motion coherence for T1-Low (black) and T1-High (red) configurations (n = 7; Repeated measures ANOVA showed a main effect for coherence and for targets configuration: <i>F(10</i>,<i>60) = 54</i>.<i>53</i>, <i>p<0</i>.<i>001</i> and <i>F(1</i>,<i>6) = 7</i>.<i>91</i>, <i>p<0</i>.<i>05</i>, respectively). Solid lines are fits to the data with a psychometric function (<i>R = 0</i>.<i>99</i> for T1-Low and <i>R = 0</i>.<i>98</i> for T1-High, <i>p < 0</i>.<i>001</i> in both cases; one-sample Wilcoxon signed-rank test, <i>p<0</i>.<i>05</i>; see <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0144841#sec002" target="_blank">Materials and Methods</a>). Error bars are SEM. <b>(B)</b> Individual subjects probability of selecting T1 for left (Left panel) and right (Right panel) for a specific level of motion coherence, solid black circles for 0%, 2%, 5% and 10% amount of motion coherence and solid gray circles for 20% and 100%. Consistent with (A), the majority of the dots are above the diagonal.