blog

When we open our eyes, the images we see are processed and represented by a collection of mulitple ‘visual areas’ at the back of our brain. Visual neuroscientists can locate many of these with ease, but the peculiar organization and location of V4 – the brain’s fourth visual field – an area involved in color vision and shape processing, has made it difficult to know exactly where this area begins and where it ends.

 Jonathan Winawer and Nathan Witthoft published an article in the INCF Gateway on F1000Research, which recently passed peer review, describing a step-by-step tutorial to systematically identify V4 in humans. They explain why localizing V4 has been such a headache, their solution to demarcating this area, and why they decided to publish this all as a protocol. 

The visual cortex is tiled with maps. Each one represents some or all of the visual field, and like nations, each occupies a different location and has its own unique characteristics. V1, or the first visual field map, was first identified in the 19th century by linking visual field loss to injuries to the posterior part of the occipital lobe. Its anatomical boundaries and representation of the visual field can now be routinely localized with a high degree of precision using standard, non-invasive neuroimaging techniques.

Retinotopic mapping with fMRI, which involves stimulating a subject’s retina with sweeping bars or circular and angular images, is most common, but the location and organization of V1 can also be inferred from anatomical features such as the pattern of sulci (grooves) and gyri (ridges) in the occipital lobe. The reliable identification of V1 and other nearby visual field maps (V2 and V3) across labs and methods has served as a building block for many research questions, including how the early visual system processes images, and how the human visual cortex compares to that of other primates.

Mapping our brain’s elusive fourth visual field

Given the interest in this area and the wide array of measurement tools available to the visual neuroscientist, why has it proven so hard to reach consensus on its location and topography?

In contrast, the fourth visual map has proven to be more elusive, with a wide variety of claims about where it is, what its neighbors are, and its topographic organization of the visual field. At the same time, there has been great interest in V4 since its discovery several decades ago, first in monkeys and later in humans. The interest in this map comes from its many purported functions, including an important role in seeing color and form and in visual attention. Given the interest in this area and the wide array of measurement tools available to the visual neuroscientist, why has it proven so hard to reach consensus on its location and topography?

The first problem in identifying a map is knowing where to look for it. V1, for example, always lies along an easily identified deep groove called the calcarine sulcus, making the map easy to find. Such a relationship was only pointed out relatively recently for V4. Nathan Witthoft in Kalanit Grill-Spector’s lab at Stanford showed that the peripheral boundary of V4 nearly always lies in the posterior transverse collateral sulcus, positioned on the ventral occipital cortex, slightly lower and to the side of the calcarine sulcus.

Second, the topography in and around V4 is complicated: V4 and its ventral occipital neighbors lie at the boundary of two map clusters, and retinotopic features twist at cluster boundaries. This contrasts with V1-V3’s long, relatively straight, iso-angle and iso-eccentricity lines (2D features of the retinotopic maps).

Third, the V4 map in many people lies near a major fMRI artifact, a large cavity that drains blood from the back of the head called the transverse venous sinus, which can obscure measurements. This artifact often affects a chunk of V4, but not the entire map, so we can at least find part of the map in all subjects.

Fourth, V4 is not as strongly and reliably driven by simple spatial patterns as V1-V3, so that the retinotopic features tend to be noisier. Finally, there has been a question about whether human V4, claimed by several groups to lie in a single region on the ventral surface, is homologous to macaque V4, which is divided into a ventral and dorsal part.

Combining and sharing knowledge

The advantage of a tutorial is that it increases the likelihood that different groups (should they use it) will define the map in the same way.

Despite the challenges described above, there has recently been a greater- albeit not complete -consensus about where V4 is and how it is organized. We decided that it would be valuable to combine the recent observations about sulci, sinuses, and retinotopy into a tutorial on how to identify the boundaries of this map.

Our tutorial – both text and video – walks through the procedure of identifying the V4 map in several example subjects, describing the heuristics we use to find the boundaries and the pitfalls we try to avoid. The advantage of a tutorial is that it increases the likelihood that different groups (should they use it) will define the map in the same way. This in turn makes it more likely that the various V4 findings reported across the visual neuroscience community will be synergistic, and help to further our understanding of how this part of the brain contributes to seeing.