Factors affecting the predictability of pseudo-random motion stimuli in the pursuit reflex of man

G. R. Barnes, C. J S Ruddock

    Research output: Contribution to journalArticlepeer-review


    Experiments have been performed on human subjects to determine the principal mechanisms underlying the break-down in performance during ocular pursuit of pseudo-random target motion stimuli composed of a mixture of two, four or six sinusoids. As observed in a previous experiment there was a reduction in the ratio of eye velocity to target velocity (eye velocity gain) for lower-frequency components of the stimulus whenever the highest frequency exceeded 0.4 Hz, but the following effects were also observed. Using a combination of four sinusoids in which the three lowest frequencies (0.11, 0.24 and 0.37 Hz) and a constant peak velocity (3 or 6 deg/s) it was shown that an increase in the velocity of the highest frequency (0.78 or 1.56 Hz) caused a progressive decline in gain of the low frequencies and a significant reduction in phase lag for the highest-frequency component. Using a combination of two sinusoids (0.44 and 1.56 Hz), in which the peak velocity was varied over a wide range (4-32 deg/s), it was shown that the reduction in low-frequency gain was dependent on the velocity ratio between the frequency components rather than their absolute velocity. Experiments using a combination of either four or six sinusoids in which the two highest frequencies were very close have revealed a true enhancement in the gain of the highest-frequency component in relation to other frequency components of the stimulus. In the same experiments the phase relationships in the response were shown to vary according to the frequency range of the stimulus in such a way that phase advance was normally present at the lowest frequency even when this ranged up to 0.89 Hz. When the oculomotor system was passively stimulated by allowing the subject to fixate a tachistoscopically illuminated stationary target, pseudo-random target motion induced a response which exhibited characteristics similar to those of active pursuit; that is, enhancement of the gain of the highest frequency and phase advance at the lowest frequency. During passive stimulation the changes in gain of the low frequencies with increasing frequency of the highest-frequency component were not consistent with those of active pursuit. However, increasing the velocity of the highest-frequency component to simulate the retinal velocity error conditions of normal active pursuit caused a significant decrease in low-frequency gain and a subjective effect of high-frequency dominance similar to that observed during active pursuit. The results indicate that the characteristic features of break-down in pursuit such as low-frequency gain reduction, highest-frequency gain enhancement and phase optimization, are inherent in the visual feed-back mechanisms and not dependent on volitional control. The results have been interpreted in terms of the behaviour of a non-linear model of the pursuit reflex, in which it is assumed that the dynamic characteristics of the visual feed-back are controlled by a secondary mechanism sensitive to the frequency and velocity content of the stimulus components. This model has been used to simulate a number of the important changes in gain and phase obtained experimentally.
    Original languageEnglish
    Pages (from-to)137-165
    Number of pages28
    JournalJournal of Physiology
    Publication statusPublished - 1989


    Dive into the research topics of 'Factors affecting the predictability of pseudo-random motion stimuli in the pursuit reflex of man'. Together they form a unique fingerprint.

    Cite this