Introduction

The classical view of cerebral cortex organization is that neocortex is subdivided into three functional types of cortex: sensory, motor, and association cortex (see Masterton & Berkley, 1974). Textbook pictures reflecting this view typically portrayed a primary area and a secondary sensory area for vision, hearing, and touch, a primary and often a secondary motor area, and association cortex (e.g., Thompson, 1967). All mammals were thought to have these sensory and motor areas but to differ in the amount of association cortex. Thus, rats, cats, monkeys, and humans were distinguished from one another by having progressively larger brains with proportionately more and more association cortex.

Over the past 30 years, this classical view has been repeatedly challenged in a number of ways. For example, this “ladder of levels” approach to describing brain differences and evolution is now seen as too simple and not reflecting the great divergences of brain organization that can be demonstrated experimentally (see Diamond & Hall, 1969; Hodos & Campbell, 1969; Preuss, 2000). Perhaps more important, regions of cortex once assigned to association cortex were rapidly being reassigned to the sensory and motor categories (see Merzenich & Kaas, 1980). In New World and Old World monkeys, as many as 35 visual areas were proposed (Felleman & Van Essen, 1991; Kaas, 1989) in regions of occipital, temporal, and parietal cortex that were formerly considered association cortex. Likewise, monkeys were found to have at least seven representations of the body (Kaas, Jain, & Qi, 2001) and approximately 12 auditory fields (Kaas & Hackett, 2000). In addition, primates appeared to have at least ten motor fields (Wu, Bichot, & Kaas, 2000). Subtracting these sensory and motor fields from the cortical sheet leaves little of the original vast expanse of association territory in these mammals. Thus, association cortex was transformed into sensory cortex (Kaas, 1999). As sensory areas of the same modality and motor areas are grouped in the same region of cortex, they could be efficiently interconnected with short axon pathways (Young, Scannell, Burns, & Blakemore, 1994). The new view that emerged from this evidence was that processing is largely unimodal, involving an array of closely interacting, specialized cortical areas, and that little tissue is devoted to multisensory integration.

Although there is much to be said in support of this new view, it now seems that it is too extreme, for several reasons. First, a number of bimodal or multisensory areas have been identified, and they have connection patterns that are consistent with this classification. Second, more detailed studies of connection patterns, as well as more physiological data, suggest that little cortex is truly unimodal. Areas long considered to be unimodal may have inputs reflecting other modalities, and neurons in most sensory and motor areas may be under the influence of more than one modality. Thus, it may be more relevant to characterize cortical areas not by their dominant modality but by the relative weights and roles of different types of inputs.

In this chapter we describe some of the areas in the neocortex of monkeys and other primates that have connection patterns that relate them to more than one modality. These few examples serve to illustrate the importance and even the dominance of bimodal and multisensory processing in the cortex of primates. Some of the proposed subdivisions of sensory and motor cortex in macaque monkeys are shown in Figure 17.1.

Figure 17.1.  

Some subdivisions of neocortex in macaque monkeys. Cortex from the intact brain (upper left) has been removed and flattened so that the fissures are open. The cortex normally hidden in fissures is gray. Multimodal or bimodal areas of cortex include the rostral region of the superior temporal sulcus (STPr), the anterior, lateral, ventral, and medial areas of the intraparietal sulcus (AIP, LIP, VIP, and MIP), and the parieto-occipital area (PO). Also shown are the caudomedial area (CM) of the auditory belt, the parietal ventral (PV) and ventral somatosensory (VS) areas, and the ventral premotor area (PMV). Other areas include primary visual cortex (V1, or area 17), the middle temporal visual area (MT), primary auditory cortex (A1), the rostral auditory area (R), the auditory belt of secondary areas (Belt), the auditory parabelt (Parabelt), posterior parietal areas 5a, 7a, and 7b, retroinsular cortex (Ri), anterior parietal somatosensory areas 3a, 3b, 1, and 2, primary motor cortex (M1), dorsal premotor cortex (PMD), the supplementary motor area (SMA) and its eye field (E), and the frontal eye field (FEF). Fissures include the cingulate sulcus (CgS), the principal sulcus (PS), the central sulcus (CS), the lateral sulcus (LS), the superior temporal sulcus (STS), the intraparietal sulcus (IPS), the lunate sulcus (LUS), the inferior occipital sulcus (IOS), the occipital temporal sulcus (OTS), and the calcarine sulcus (CAS). The pyriform cortex (PY) and the corpus callosum (CC) are identified.