Upcoming Events

February 27, 2023
12:00 p.m., URMC Case Methods Room, 1-9576

Identifying Targets in Noise and Natural Backgrounds

This talk will describe evidence for a theory of detection in natural scenes based directly on the statistical properties of natural images, and on the variation in retinal processing with retinal location. We have identified five approximately-independent properties of natural backgrounds that contribute substantially to the detectability of achromatic targets in the fovea: the local background luminance, the local background contrast, the statistical uniformity of the background under the target, the phase-invariant similarity of the target and background, and the phase-dependent similarity of the target and background. The lawful effects of these five properties are quantitatively predicted, with almost no free parameters, from natural scenes statistics, together with a direct measurement of intrinsic position uncertainty. This occurs because of the relatively constant efficiency of cortical processing. The variation in threshold with retinal location appears to be largely explained by the variation in ganglion-cell-sampling density with retinal location, together with the corresponding variation in intrinsic position uncertainty, although this is still a work in progress. A better understanding of detection in natural scenes is a crucial step toward understanding the more natural task of visual search in natural scenes.

---

March 2, 2023
4:00 p.m., URMC 3-7619 Upper Auditorium

---

April 20, 2023
4:00 p.m., URMC 3-7619 Upper Auditorium

---

May 3, 2023
3:00 p.m., URMC 2-6408, K207 Auditorium

Neural dynamics of the primate attention network

---

May 17, 2023
3:00 p.m., URMC 2-6408, K207 Auditorium

Neurophysiological windows onto decision formation in the human brain

---