DETECTION OF CHANGES IN SPEED AND DIRECTION OF MOTION - REACTION-TIMEANALYSIS

Observers reacted to the change in the movement of a random-dot field whose initial velocity, V-0, was constant for a random period and then switched abruptly to another value, V-1. The two movements, both hori zontally oriented, were either in the same direction (speed increments or decrements), or in the opposite direction but equal in speed (dire ction reversals). One of the two velocities, V-0 or V-1, could be zero (motion onset and offset, respectively). In the range of speeds used, 0-16 deg/sec (dps), the mean reaction time (MRT) for a given value of V-0 depended on \V-1 - V-0\ only: MRT approximate to r + c(V-0)/\V-1 - V-0\(beta), where beta = 2/3, r is a velocity-independent component of MRT, and c(V-0) is a parameter whose value is constant for low valu es of V-0 (0-4 dps), and increases beginning with some value of V-0 be tween 4 and 8 dps. These and other data reviewed in the paper are acco unted for by a model in which the time-position function of a moving t arget is encoded by mass activation of a network of Reichardt-type enc oders. Motion-onset detection (V-0 = 0) is achieved by weighted tempor al summation of the outputs of this network, the weights assigned to a ctivated encoders being proportional to their squared spatial spans. B y means of a ''subtractive normalization,'' the visual system effectiv ely reduces the detection of velocity changes (a change from V-0 to V- 1) to the detection of motion onset (a change from 0 to V-1 - V-0). Su btractive normalization operates by readjustment of weights: the weigh ts of all encoders are amplified or attenuated depending on their spat ial spans, temporal spans, and the initial velocity V-0. Assignment of weights and weighted temporal summation are thought of as special-pur pose computations performed on the dynamic array of activations in the motion-encoding network, without affecting the activations themselves .