Models of Response Time to Peripheral Stimuli

Abstract

The variable contribution of peripherally presented stimuli in a A sensory motor task has been explored in terms of stimulus and environmental variables. A simulated driving task was chosen as being a representative compensatory tracking task.

Empirical models have been developed using response surface methodology, statistical design and data collected on a simulator with a 240° wrap-around screen and projection systems very much like cinerama. In this research, seven factors were isolated for a study of their effects on detection latency to peripherally presented stimuli when the subject was ‘driving’. These factors were stimulus size (circular stimuli between 18′ and 60′), stimulus color (red, white and green), stimulus-background contrast (background luminance 1ft.L and stimulus luminance of 30, 60 and 90 ft.L), stimulus location along the horizontal (between ± 90°) and vertical meridians (between ± 26°), intensity of continuous white noise (between 52 and 100 dbA), and complexity of the continuous central tracking task measured in terms of the simulated driving speed. Three levels of each variable were selected in a 7 factor Box-Behnken design. Twenty undergraduates between the ages 19 and 26 participated in the experiment.

It was found that, in this multivariable environment when all seven factors were simultaneously varied, the effects of noise, stimulus location in the visual field and stimulus size were the more important determinants of response latency. In addition, marked differences for the left and right visual fields were observed for the right-handed subject population.

Four models have been developed: two for the left visual field, with and without the continuous central task (CCT), and two for the right visual field for the same conditions. The response was found to be of the form 1/Yr = f (x_i); i= 1,2,… 7 for both the left and right visual fields in the presence of the CCT. In the absence of the CCT the model was of the form Y_r = f (x_r) for the left and 1/2 = f (x_i) for the right visual field where Y_r = response time in millisec. and Y_r x_i = variables in equations.

Response curves have been presented to illustrate the variation of response time with each of the seven variables for regions where response time may be expected to be a minimum. The implications of these curves and the models on which they are based have been examined from the design point of view.