The eyes, head, and body are often in motion, and many of the most relevant objects in the environment are those that move (e.g., other organisms). Furthermore, motion provides an important source of information about distance and shape. Thus, the perception of motion is a critical function of the visual system. For a number of years, my laboratory in collaboration with Albrecht’s laboratory has been investigating mechanisms underlying the perception of motion direction (Hamilton, et al., 1989; Albrecht & Geisler, 1991; 1994; Geisler & Albrecht, 1997). Most recently, we have been investigating the contribution of spatial mechanisms to the perception of motion direction. When an image feature moves quickly it should create a spatial streak in the direction of the motion due to temporal integration in visual system. Thus, motion blur should provide a purely spatial signal for the direction of motion. We have evidence from psychophysical experiments and that such signals exist and are used by the human visual system (Geisler, 1999). We have also shown in neurophysiological experiments that such signals exist in the primary visual cortex of monkey and cat, and that the properties of these signals are consistent with the human psychophysical results (Geisler, et al., in press). One of the neurophysiological experiments is illustrated in the figure on the left. A symmetrical spot is moved either parallel or perpendicular to the flanks of the receptive field of a neuron in primary visual cortex. When the spot moves slowly the response is on average a bit larger for perpendicular motion. However, when the spot moves quickly, temporal integration creates an elongated spatial streak and hence the response becomes larger for parallel motion. The results of a psychophysical experiment are shown in the figure on the right. Thresholds for detecting a moving spot were measured when a superimposed dynamic random line mask was oriented either parallel or perpendicular to the direction of the spot motion. At slow speeds the ratio of the thresholds for parallel and perpendicular masks were nearly 1.0, but above a critical speed of 10-12 spot-widths/sec the thresholds were greater for the parallel mask. A critical speed of 10-12 spot-widths/sec was also observed in the responses of neurons in primary visual cortex. For small features, this critical speed is less than 1 deg/sec, which is quite slow. Our studies suggest that motion streaks are one of several cues the visual system uses to determine motion direction.