We studied differences in the development of sensitivity to local versus global motion by comparing the thresholds of 5-year-olds and adults (n = 48/age) tested with stimuli that moved at slower (1.5 /sec) or faster (6 /sec) speeds. To measure sensitivity to local motion, we used horizontal sine-wave gratings that were luminance-modulated (first-order) or contrast-modulated (second-order) and varied the amount of modulation over trials. To measure sensitivity to global motion, we used Random-Gabor Kinematograms (RGKs) formed with luminance-modulated (first-order) or contrast-modulated (second-order) concentric Gabor patterns equated for visibility. The percentage of dots moving coherently varied over trials. The task on each trial was to say whether the stripes (local motion) or Gabors (global motion) were moving up or down. We calculated individual immaturity scores for each condition by comparing a child’s threshold to the mean threshold for adults tested under that condition. Children’s thresholds were immature for all conditions (ps < .0001). Mean immaturity scores ranged from as much as 0.96 log units for second-order global motion at 1.5 deg/sec to as little as 0.02 log units for first-order local motion at 6 deg/sec. An ANOVA on the immaturity scores revealed a 3-way interaction between speed, modulation type, and motion type. At both speeds, immaturity scores were larger for global motion than for local motion and larger for second-order modulation than for first-order modulation. Thus, aspects of motion processing deemed to require more steps of cortical processing, namely global motion and second-order motion, are slower to mature during childhood than those deemed to require fewer steps of cortical processing, namely local motion and first-order motion.

Supported by Canadian Institutes of Health Research grant # MOP-36430