Phase Encoding

Phase Encoding

Visual input, coming from the retina, is decomposed by the cells in the primary visual cortex (V1) that are tuned to different spatial frequency, orientation and phase. A number of studies have argued for a phase model involving detectors tuned to two or more of four channels. The project I have done is to investigate whether there exist detectors of four cardinal phases in human vision. We found that only two phase mechanisms, ±cosine, in human vision.

However, not many studies investigate how phases integrates across different spatial scales (e.g. spatial frequency) and how the phase information influence our identification of images. Using complex stimuli in the future studies would help to answer the phase encoding mechanism.

• Huang, P.C., Kingdom, F. A. A. & Hess, R. F. (2006). Only two phase mechanisms, ±cosine, in human vision. Vision Research, [Abstract] 46(13):2069-2081.

Relative phase between frequency components is also important for investigating how visual system processes complex images. If the relative phases between components are aligned with each other, they can form features in the natural scene. Combining with pattern masking paradigm, I investigated how phase-aligned and phase-scrambled stimuli (with the same contrast energy) influenced our detectability of the target. The results showed that features caused much stronger masking effect .

• Huang, P.C., Maehara, G., May, K. A. & Hess, R. F. (2012). Pattern masking: the importance of remote spatial frequencies and their phase alignment. [Abstract] J Vis, 12(2):14, 1-13.