Purchase this article with an account.
Benjamin Harvey, Alan Cowey, Oliver Braddick; Similar processing for detection and position discrimination of expanding, contracting and rotating motion flow patterns in random dot kinematograms, shown by adaptation and TMS. Journal of Vision 2008;8(6):16. doi: 10.1167/8.6.16.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Accurately localizing the centre of an expanding, contracting or rotating motion pattern is a key element of optic flow processing, necessary for example for deriving heading from optic flow. We have previously shown that, unlike the detection of an optic flow pattern, centre-of-motion localization depends on a very small region near this centre (Harvey and Braddick, 2006 Perception 35S, 238). Together with studies showing normal localization in a stroke patient with direction discrimination deficits (Beardsley and Vaina, 2005 Journal of Computational Neuroscience 18 55) and retinotopic activation of V3A depending on the positions of centres of radial motion (Koyama et al, 2005 Current Biology 15 2027–2032), this suggests that centre localization and flow pattern detection may depend on distinct cortical systems. Here we show, however, that adaptation to radial or rotating flow patterns affects coherence thresholds for detection and centre-of-flow position discrimination similarly, and at a global processing level. For both pattern detection and position discrimination, adaptation to the tested pattern or to a flow pattern moving in the opposite direction (e.g. contracting adaptation for an expanding test pattern) increases test coherence thresholds. In both tasks, however, adaptation to expansion does not significantly impair processing of contraction in most subjects. In neither task did adaptation to a rotating pattern affect performance on a radial test pattern, or vice versa. Furthermore, repetitive TMS stimulation of hMT+ affects both tasks, while TMS over V3A and nearby areas affects neither. These results suggest that the position of motion patterns is encoded in the global motion processing areas of hMT+. Furthermore, we show that when detection and position discrimination tasks are equated for difficulty, pattern detection can be achieved after a briefer exposure than position discrimination, suggesting that accurate position information requires longer temporal integration than detection of a flow pattern.
This PDF is available to Subscribers Only