MIT OpenCourseWare: Visual Motion Processing, MT Area and Brain Connectivity

Overview

This MIT OpenCourseWare lecture examines motion perception, cortical maps, and the brain circuits that support everyday vision. It covers how motion information travels from the eyes through the thalamus to the cortex, the special role of area MT in motion processing, and how receptive fields and retinotopy create organized maps in visual cortex. The talk also touches on memory and emotion structures such as the hippocampus and amygdala, and demonstrates how high level computations and connectivity shape perception.

Key insights

• The thalamus acts as a major gateway for sensory input to cortex, with LGN relaying visual information to visual cortex.
• Area MT contains neurons tuned to motion direction and forms a topographic map of motion preferences across cortex.
• Receptive fields define where in the visual field a neuron responds, and neighboring cortical neurons share similar receptive fields, creating retinotopic maps.
• Memory and emotion structures like the hippocampus and amygdala influence perception and behavior, illustrating the brain’s integrated approach to sensing and interpretation.

Introduction and the Brain’s Motion Processing

The lecture frames motion as a fundamental cue that animals and humans rely on for ecological interaction, navigation, and social behavior. It begins with a broad neuroscience primer, emphasizing that the cortex forms a large, folded sheet responsible for higher processing, and that outside of the cortex lie critical subcortical structures and white matter tracts that connect regions.

Neuroanatomy and Core Concepts

The speaker outlines four major brain components: the brainstem, cerebellum, limbic/subcortical regions, and the cortex. The thalamus is highlighted as a central relay that channels sensory information from peripheral organs to cortical areas; the LGN is the specific thalamic relay for vision. The hippocampus and amygdala are introduced as key subcortical structures involved in memory and emotion, respectively, with real patient examples illustrating their roles. White matter is discussed as bundles of axons enabling long-range connectivity, forming the brain’s connectivity fingerprint which helps distinguish different cortical areas.

Primary Cortical Maps and Receptive Fields

Primary sensory regions (visual, auditory, somatosensory, gustatory, and motor) each host maps that reflect putative sensory spaces. Receptive fields are defined as the regions of the visual world that drive a given neuron’s firing. The map of visual space in the cortex, termed retinotopy, shows that adjacent cortical areas correspond to adjacent parts of the visual field. The concept underpins the systematic organization of the visual cortex, including V1 (striate cortex) and neighboring areas.

Area MT and Motion Processing

The core of the talk focuses on area MT, a visual area specialized for motion processing. In monkeys, MT neurons show direction selectivity, responding preferentially to motion in certain directions. Across MT, there is a coherent map of direction preference analogous to retinotopy in V1. In humans, MT can be studied with functional MRI, which reveals motion sensitivity in MT when subjects view moving versus stationary dots. While fMRI shows motion sensitivity, inferring direction selectivity at the single-neuron level in humans relies on indirect methods such as psychophysical aftereffects and brain stimulation studies.

Psychophysics and Aftereffects

The lecture demonstrates an aftereffect used to infer directional tuning in human motion processing. By presenting a prolonged motion stimulus and then observing perceptual bias when viewing stationary stimuli, the speaker explains how adapting sensory neurons can reveal direction-selective populations without invasive recording. This psychophysical approach complements invasive single-neuron recordings in animals and human brain imaging.

Structure, Connectivity, and Cortical Areas

The talk concludes by revisiting the criteria for identifying cortical areas: distinct function, unique connectivity fingerprints, and sometimes distinct cytoarchitecture. MT is highlighted as a robust exception to clear cytoarchitectural boundaries, with clear histological and functional demarcation. The connectivity map shows MT’s integration with other visual areas, supporting its role in motion perception. The lecturer emphasizes that understanding perception requires considering ecological purposes and the computational challenges of inference, linking Marr’s framework to contemporary neuroscience.