The broadest inclusion criteria used in the literature for prefrontal cortex is the receipt of projections from the DORSOMEDIAL NUCLEUS OF THE THALAMUS. In the brain of PRIMATES, the dorsomedial nucleus of the thalamus (MD) sends projections to two areas that have been called classic prefrontal cortex, namely, DORSOLATERAL PREFRONTAL CORTEX and ORBITOFRONTAL CORTEX. ANTERIOR CINGULATE CORTEX also receives dorsomedial nucleus projections. In the primate brain, frontal cortex can be divided into areas with a granular layer 4 and agranular areas (see AGRANULAR CORTEX; CORTICAL LAYERS). Prefrontal cortex has also been defined as the granular areas of the primate FRONTAL LOBE, which would include dorsolateral prefrontal cortex and orbitofrontal cortex but may exclude anterior cingulate, which has a less well developed granular layer.
In the RAT brain, there is no area of FRONTAL CORTEX with a granular layer, such that the definition of prefrontal cortex is more problematic. Areas with dorsomedial nucleus projections are on the MEDIAL WALL (from SHOULDER CORTEX [M2] to INFRALIMBIC CORTEX and, on the ventral surface, VENTRAL ORBITAL CORTEX [VO], an area related to the orbitofrontal cortex of the primate brain). The projections to infralimbic cortex are sparse and from the only the most medial portion of the dorsomedial nucleus, leading to some querying the inclusion of this area. Dorsally, M2 is thought to be homologous to secondary motor cortex of the primate and therefore not prefrontal. Ventral to M2, areas Cg1 and 2 (see CG1/2/3), and possibly PRELIMBIC CORTEX, may be homologous to primate anterior cingulate cortex and not, therefore, classical prefrontal cortex.
There is some confusion about the function of prefrontal cortex, due to several factors. There is the most obvious problem of defining the extent of prefrontal cortex in the primate brain. There is reasonably good agreement about the inclusion of some areas (for example, dorsolateral prefrontal cortex). This problem mainly concerns anterior cingulate cortex, which some authors include as prefrontal cortex and others group as LIMBIC CORTEX, acknowledging the projection from dorsomedial nucleus but citing the less well-developed granular layer in primate anterior cingulate. Given that there is no consensus about the boundaries of prefrontal cortex in the primate brain, let alone the location of homologous areas in the brains of other species, it is difficult to make cross-species comparisons. This problem is compounded because the functions of prefrontal cortex are complex, so that there is difficulty in designing tasks suitable for making comparisons between species and demonstrably measuring equivalent function. In addition, the complexities of species-specific cognitive abilities means that at least some of the functions of prefrontal cortex may have to be defined in species-specific terms (LANGUAGE and verbal reasoning being the most obvious examples). Finally, as different regions of prefrontal cortex appear to process different kinds of information, proposing a single function of prefrontal cortex might be too general to be empirically useful.
The term EXECUTIVE FUNCTIONS is often applied to the functions of the prefrontal cortex, but this term is usually defined so broadly as to have limited practical use in predicting deficits following lesion damage of specific regions of prefrontal cortex. Furthermore, in its broadest use, the term executive function is applicable to other brain areas and therefore is not useful as a definition of prefrontal cortex. The alternative to using such an embracing term is to list the individual tasks on which impairments are found. This approach is inherently unsatisfying: it might be that task complexity per se is the reason for impairments in performance. Alternatively, if the task depends upon several fundamental cognitive abilities, performance impairments might indicate a problem with any one of a variety of different functions. Furthermore, an impairment of some fundamental cognitive operation might result in impairments on seemingly disparate tasks.
Nevertheless, with this caveat, it is possible to specify tasks on which performance impairments associated with damage to prefrontal cortex are commonly reported. For example, performance of tasks that involve making a spatial discrimination based on information presented at some prior time (for example, SPATIAL MEMORY) is impaired when the area around the PRINCIPAL SULCUS is functionally compromised. Damage to the cortex between the principal sulcus and the ARCUATE SULCUS impairs any delayed response task (that is, tasks involving WORKING MEMORY), even without a spatial component. Holding different kinds of information ‘on-line’ in working memory would seem to be an important requirement for anticipating outcomes, response planning and initiating goal-directed behaviour. Thus it is reasonable to suggest that a working-memory deficit following damage to prefrontal cortex is fundamental, with implications for a variety of complex behaviours. Nevertheless, impairments in ATTENTIONAL SET-SHIFTING following lesions of dorsolateral prefrontal cortex cannot be explained by a working-memory deficit, so either there is another, yet more fundamental, deficit or there must be multiple deficits.
