The purpose of this project is to provide evidence to support the prediction impairment hypothesis (PIH) of autism (Sinha et al., 2014, PNAS). This is significant because, thus far, an underlying, global explanation of characteristic behavior in autism has not been found. The PIH provides an exciting hypotheis that needs further evidence. 16 children with ASD and 16 age- and IQ-matched neurotypical (NT) children participated in the experiment. They were asked to catch a ping-pong ball launched from a ball machine. 3D motion capture recorded joint kinematics of participantsÕ arms and the flying ball. A set of control tasks assessed postural balance, reaction time when lifting both arms, and movement speed in reaching to a stationary ball. We hypothesized that in these control tasks, ASD and NT participants would not differ. Results showed that ASD participants indeed had fewer successful catches than NTs (p=0.01). Detailed kinematic analyses of ball and hand trajectories revealed that while NTs approached the ball with a relatively stable profile of hand velocity, ASDs exhibited a collision-like motion, suggesting insufficient prediction of the upcoming ball. In the control tasks, there was a lack of differences in basic aspects of motor coordination which support the interpretation that it is the predictive demands that are likely responsible for the reduced performance of ASD participants when catching a flying ball.