Purchase this article with an account.
Jürgen Kornmeier, Rike Wörner, Michael Bach; When our brain is convinced: EEG correlates of visual ambiguity. Journal of Vision 2014;14(10):1239. doi: 10.1167/14.10.1239.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
During the observation of an ambiguous figure, like the Necker cube, our perceptual system is unstable and alternates spontaneously between two (or more) mutual exclusive representations. Tiny figural changes can disambiguate the ambiguous figure, thus stabilize one representation and induce two sizable ERP components, a fronto-central P200 and a parieto-central P400. These ERPs are absent with ambiguous stimulus variants. This pattern of results was found with geometric (Necker lattice) and semantic (Old/Young woman) stimuli. In the current study we looked (1) for similar effects with ambiguous and unambiguous motion stimuli and (2) for correlations between the degree of ambiguity and the amplitude of the two ERP components. Methods: 12 Participants (six females) viewed ambiguous SAM (stroboscopic alternative motion) and Necker lattice stimuli and variants with different degrees of disambiguation in separate experimental blocks. Stimuli were presented discontinuously (≈ 800 ms) with short blank-screen intervals (≈ 20 ms) and participants manually indicated perceptual alternations between presentations. EEG traces were sorted for stimulus category (geometry, motion) and degree of ambiguity and averaged to ERPs. Latencies and amplitudes of P200 and P400 were analyzed. Results: (1) Tiny figural changes, rendering an ambiguous figure unambiguous, evoke two sizable positivities at about 200 and 400 ms after stimulus onset ("P200/P400 Ambiguity Effect"), which are absent with the ambiguous figures. (2) The two ERP amplitudes increases linearly with decreasing ambiguity of the stimulus. (3) This P200/P400 Ambiguity Effect was replicated for the Necker lattices and also found with SAM motion stimuli. Our results suggest the existence of a neural evaluation instance that estimates the reliability of the perceptual outcome, given limited and ambiguous visual input. This evaluation seems to work on a processing level generalized across sensory domains.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only