We cordially invite you to the next lecture of the BME Cognitive Seminar Series:

Date & Time: October 29, Monday, 12:00-13:00
Location: BME, XI., Egry József utca 1., T. ép 515.

Sensory noise processing in the human brain: insights from object recognition studies on healthy and amblyopic subjects

Bankó Éva
Information Technology Department, Pázmány Péter Catholic University
Neurobionics Research Lab

http://vision.itk.ppke.hu/
Personal webpage: http://digitus.itk.ppke.hu/~banko/

Abstract
Human visual object recognition is fast and efficient when viewing conditions are good. However, under low visibility conditions the visual system must recruit additional processing resources to handle the noisy and deteriorated visual images, thus object recognition becomes more effortful. Even though this is often the case, little is known about the neural mechanisms engaged in processing of noisy images. This is even more important since adding noise to images is widely used in decision-making studies to modulate task difficulty while not taking into account the confound introduced by the increased sensory processing demands.  In a series of experiments on healthy and amblyopic subjects we have pinned down both in time and space the active processes associated with sensory noise processing using faces with decreased phase-coherence. Namely, phase noise affects the electrophysiological responses in the first 300 ms following stimulus onset that is unrelated to changes in task-difficulty; most importantly, there is an increase in the ERP single-trial (i.e. true) response amplitudes between 200 - 300 ms after stimulus onset – involving the P2 component – the noise-modulation of which is diminished in amblyopia. This amblyopic deficit measured on the P2 component predicted the severity of the noise-related behavioral impairments and could not be explained by an overall increase in stimulus uncertainty or task difficulty in the case of noisy stimuli, which have been also proposed previously as explanations of the noise-related ERP changes. On the other hand, the noise-modulation of the N170 component of the ERP responses – reflecting structural processing of face images – was similarly affected by the presence of noise in the amblyopic and the fellow eye of amblyopes, suggesting that the noise-induced decrease of the N170 could simply be due to the decrease in the face content of the images. Furthermore, we also showed that processing of phase randomized as compared to intact faces is associated with increased fMRI responses in specific areas of the lateral occipital cortex. These results suggest that efficient processing of noisy images depends on the engagement of visual cortical mechanisms that take place after the early structural processing of visual objects has been completed and is reflected in the P2 component of the ERP response sand most likely localized to a retinotopic part of the lateral occipital cortex that has been implicated in grouping and image segmentation.

-- 
Attila Keresztes

Junior Research Fellow
Budapest University of Technology and Economics
Dept. of Cognitive Science,
Egry József u. 1, Budapest
1111, Hungary
Tel: +36 1 4633525