Optic flow refers to the motion pattern formed on the retina as an observer moves through the environment. Although optic flow conveys rigid 3D structure information associated with self-motion, prior studies have often conflated it with non-rigid “unreal” motion patterns that lacks coherent 3D structure, leading to ambiguity in interpretations of functional specificity across motion-sensitive visual brain regions.
Here, we combined a novel psychophysical paradigm with functional magnetic resonance imaging (fMRI) to systematically compare behavioral sensitivity and neural responses to “real” versus “unreal” optic flow. Two types of stimuli were constructed: real optic flow derived from 3D random-dot clouds, and unreal optic flow generated by temporally scrambling local velocities. Crucially, the two stimulus types were matched in two-dimensional image features and motion statistics, differing only in whether they conformed to a rigid self-motion structure.
Psychophysical results revealed that contraction patterns were more readily detected than expansion patterns in unreal optic flow, whereas the opposite pattern emerged for real optic flow. Moreover, for real optic flow, observers’ precision in localizing the center of motion significantly improved as stimulus duration increased from 100 ms to 400 ms, an effect absent for unreal optic flow. Multivoxel pattern analysis of fMRI data further showed that only neural response patterns in the dorsal medial superior temporal (MST) area mirrored these behavioral dissociations.
Together, these findings demonstrate that the visual system implements specialized mechanisms for extracting behaviorally relevant information from real optic flow, supporting accurate self-motion perception. More broadly, they suggest that area MST may function as an “expertise” region along the dorsal visual pathway, selectively tuned by the ecological statistics of everyday forward self-motion.

—

To request the Zoom link send an email to jteeters@berkeley.edu.  Also indicate if you would like to be added to the Redwood Seminar mailing list.

The Fitzhugh-Nagumo model with diffusive coupling is known to admit a variational (energy-based) formulation, which is a result of its underlying skew-gradient structure. We show that since stationary solutions of the diffusive Fitzhugh-Nagumo model are described by self-adjoint operators, the methods of equilibrium propagation for performing credit assignment can be applied. Further, for networks with the topology of a deep residual neural network, we show that the steady state solutions also admit a Hamiltonian description, and thus the methods of Hamiltonian Echo Backpropagation can be applied. We end by deriving an explicit layer-wise Hamiltonian recurrence relation governing inference in such models.

—

To request the Zoom link send an email to jteeters@berkeley.edu.  Also indicate if you would like to be added to the Redwood Seminar mailing list.

To be announced.

—

To request the Zoom link send an email to jteeters@berkeley.edu.  Also indicate if you would like to be added to the Redwood Seminar mailing list.

To be announced.

—

To request the Zoom link send an email to jteeters@berkeley.edu.  Also indicate if you would like to be added to the Redwood Seminar mailing list.