Search

It is now widely accepted that the prefrontal cortex (PFC) plays a critical role in the neural network subserving working memory (WM). At least three related questions are still under debate: (1) is the PFC critical for all constituent processes of WM (i.e., short-term storage, manipulation, and utilization of mental representations) or only in one or a few of them? (2) Is there segregation of function among different cytoarchitectonic subdivisions of the PFC? (3) If this be the case, is this segregation based on the nature of the information being processed or on the type of cognitive operation performed? The present review article describes findings in the monkey supporting a modular "domain-specific" model of PFC functional organization with respect to WM operations. In this model, the dorsolateral prefrontal cortex (DLPFC) is composed of several subregions, based primarily on the nature of the information being processed in WM. Storage and processing functions are integrally related in each area. Future studies designed to map as yet uncharted areas of prefrontal cortex with refined anatomical and physiological approaches may provide a critical test of the model and evaluate the extent to which it applies generally or, instead, mainly to visual domains or only to dorsolateral convexity areas.

Evidence is now converging which suggests that, at the area level, working-memory processes within the dorsolateral and ventrolateral frontal cortices are organised according to the type of processing required, rather than according to the nature (i.e. domain) of the information being processed, as has been widely assumed. In a recent study using functional magnetic-resonance imaging (fMRI), performance of visual spatial and visual non-spatial working-memory tasks was shown to involve identical regions of the lateral prefrontal cortex when all factors unrelated to the type of stimulus material were appropriately controlled. These results concur fully with recent reviews of the imaging literature, which demonstrate that spatial, visuospatial and verbal working-memory studies have produced distributed patterns of overlapping activation foci within these lateral frontal regions. Moreover, two recent positron-emission tomography studies have demonstrated that either, or both, the ventrolateral and dorsolateral frontal regions can be activated in spatial working-memory tasks, depending on the precise executive processes that are called upon by the task being performed. Similarly, when the executive requirements of a simple verbal working-memory task (e.g. forwards versus backwards digit span) are manipulated, differential activation within these two frontal cortical areas is observed. The results provide further evidence that the mid-dorsolateral and mid-ventrolateral frontal cortical areas make distinct "executive" contributions to memory and correspond with a fractionation of working-memory processes in psychological terms.

The effect of large, low thoracic (T10-T11), partial spinal lesions involving the ventral quadrants of the spinal cord and, to a different extent, the dorsolateral funiculi, on fore-hindlimb coordination was examined in cats walking overground at moderate speeds (40-100 cm/s). Three different forms of impairment of fore-hindlimb coordination depending on the extent of the lesions, were observed. Lesions sparing the dorsolateral or the ventral funiculus on one side preserved the equality of the fore- and hindlimb locomotor rhythms but changed the coupling between the movements of both girdles as compared to intact animals. Larger lesions in which, in addition to the ventral quadrants of the spinal cord, also major parts of the dorsolateral funiculi were destroyed elicited episodes of rhythm oscillations in both girdles, which appeared at the background of a small difference in these rhythms. Lesions destroying almost the whole spinal cord induced a permanent difference (about 200 ms) in the step cycle duration of the fore- and the hindlimbs. However, even in these animals some remnant form of fore-hindlimb coordination was found. The results suggest that dorsolateral funiculi play a major role in preserving the equality of rhythms in the fore- and the hindlimbs, while lesions of the ventral quadrants change the coupling between limbs.

We studied the role of the dorsolateral prefrontal cortex in procedural learning. Normal subjects completed several blocks of a serial reaction time task using only one hand without or with concurrent non-invasive repetitive transcranial magnetic stimulation. To disrupt their function transiently, stimulation was applied at low intensity over the supplementary motor area or over the dorsolateral prefrontal cortex contralateral or ipsilateral to the hand used for the test. Stimulation to the contralateral dorsolateral prefrontal cortex markedly impaired procedural implicit learning, as documented by the lack of significant change in response times during the task. Stimulation over the other areas did not interfere with learning. These results support the notion of a critical role of contralateral dorsolateral prefrontal structures in learning of motor sequences.

After WGA-HRP injections in the pontine grey involving the dorsolateral pontine nucleus, a great number of labeled cells were found in the superficial layers of the ipsilateral superior colliculus. The majority of these cells were located in the stratum griseum superficiale (SGS). Few labeled cells were found in the stratum opticum, and the stratum zonale (SZ) showed no labeled cells. Labeled cells in the SGS formed a rather homogeneous population as most of them had fusiform somata with an upper dendritic process which runs vertically to reach the SZ. These cells were mainly located in the middle third of the SGS, forming a sublamina in this layer. These results demonstrate the participation of the superficial tectal layers in the ipsilateral descending pathway of the superior colliculus, allowing visual information to reach precerebellar stations at the dorsolateral pontine nucleus.

The course and termination of afferents in the spinal dorsolateral fascicle to some medullary sensory nuclei were studied by tracing degeneration following lesions of spinal white matter. The main conclusions depend on successive degeneration experiments; other points were studied with single-stage lesions. The dorsal column nuclei were particu-larly studied; terminations in these nuclei following dorsolateral lesions followed a clear-cut pattern, with fibres arising from segments below T 6 terminating in the gracile nucleus and those with more rostral origin solely in the cuneate nucleus. In both nuclei, the major terminations were in their rostral third with most fibres traversing deep caudal regions where some termination also occurred. Some fibres ended contralaterally. These restricted regions of termination contrasted with the wide-spread terminations seen after lesions of the dorsal column. A region at the cuneate rostral pole, adjacent to but clearly separable from nucleus z, receives a dense projection from both caudal and rostral spinal levels, the former fibres terminating in the dorsal part of the region, the latter extending more ventrally. We treat this as a separate subnucleus. The afferents to the dorsal column nuclei (together with those terminating in the other nuclei studied) were confined to the extreme dorsolateral white matter. Our observations confirm the established view that only afferents arising from caudal segments (below at least T 4–5) terminate in nucleus z, and that afferents terminating in group x arise from all levels (at least between C 5 and L 5): also that neither receives any afferents through the dorsal columns. Dorsolateral fibres arising from segments above at least T 6 terminate in a clear-cut area at the lateral border of the external cuneate nucleus. Heavy terminal degeneration was also seen in the lateral cervical nucleus of afferents arising from both above and below T4–5

The rostral thalamo-hyperstriatal projection in young chicks was examined following large injections of wheat germ agglutin labelled with horseradish peroxidase (HRP-WGA) into the hyperstriatum. Retrograde labelling of thalamic neurons was present in the dorsolateral thalamus, rostrolateral part (DLAlr) and dorsolateral thalamus, lateral part (DLL). There was no evidence of a contralateral projection from the lateral anterior thalamic nucleus (LA) to the posterior aspect of the visual hyperstriatum as reported recently by Boxer and Stanford (1985). Furthermore, a comparison of labelled neurons in the contralateral rostral thalamus following injections into either the left or right hyperstriatum revealed no difference in the number of neurons. The study could therefore not confirm the presence of an asymmetrical LA-hyperstriatal projection, as reported by the above authors.

The visual properties of 77 dorsolateral pontine nucleus (DLPN) cells were studied in two alert monkeys. In 41 cells, presentation of a moving random dot background pattern, while the monkeys fixated a stationary spot, elicited modulations in discharge rate that were related either to (i) the velocity of background motion in a specific direction or to (ii) only the direction of background movement. Thirty-six DLPN cells exhibited responses to small, 0.6–1.7 deg, visual stimuli. Nine such cells exhibited non-direction selective receptive fields that were eccentric from the fovea. During fixation of a stationary bluish spot, the visual responses of 27 DLPN cells to movement of a small, white “test” spot were characterized by two components: (1) as the test spot crossed the fovea in a specific direction, transient velocity-related increases in discharge rate occurred and (2) a maintained, smaller increase in activity was observed for the duration of test spot movement in the preferred direction. This DLPN activity associated with small visual stimuli was also observed during smooth-pursuit eye movements when, due to imperfect tracking, retinal image motion of the target produced slip in the same direction. These preliminary results suggest that the DLPN could supply the smooth-pursuit system with signals concerning the direction and velocity of target image motion on the retina.

The cortical distribution of the cells of origin of the dorsolateral and the ventral corticospinal tracts were studied in cat. This was done by making subtotal spinal transections, which in different experiments spared different portions of one ventral or one lateral funiculus at C5--C7. One week later horseradish peroxidase (HRP) injections were made one segment caudal to the lesion and the cortical distribution of the HRP labeled neurons was studied. Thus, it was found that the dorsolateral corticospinal tract at C5--C7 is composed of crossed and uncrossed fibers in a ratio of about 10 : 1, while the ventral corticospinal tract, which contains much fewer cortical fibers, is composed and uncrossed fibers in a ratio of approximately 1 : 1. Further, the primary motor cortex (area 4) was found to contribute fibers to both the crossed and the uncrossed dorsolateral corticospinal tract as well as to both the crossed and the uncrossed ventral corticospinal tract. The primary somatosensory cortex (area 3a, 3b, 1--2, 5a, 5b) as well as the secondary somatosensory cortex (area 2 pre-insularis), on the other hand, were found to contribute fibers mainly to the crossed dorsolateral tract. Area 4 was found to display a further organization, such that it contains a medial and a lateral part, both of which contribute mainly fibers to the crossed dorsolateral tract, while the remainder of area 4 contributes fibers to the crossed and uncrossed dorsolateral as well as to the crossed and uncrossed ventral tracts.

This paper reports evidence linking dorsolateral prefrontal cortex with one of the cognitive abilities that emerge between 7.5-12 months in the human infant. The task used was Piaget's Stage IV Object Permanence Test, known as AB (pronounced "A not B"). The AB task was administered (a) to human infants who were followed longitudinally and (b) to intact and operated adult rhesus monkeys with bilateral prefrontal and parietal lesions. Human infants displayed a clear developmental progression in AB performance, i.e., the length of delay required to elicit the AB error pattern increased from 2-5 s at 7.5-9 months to over 10 s at 12 months of age. Monkeys with bilateral ablations of dorsolateral prefrontal cortex performed on the AB task as did human infants of 7.5-9 months; i.e., they showed the AB error pattern at delays of 2-5 s and chance performance at 10 s. Unoperated and parietally operated monkeys succeeded at delays of 2, 5, and 10 s; as did 12 month old human infants. AB bears a striking resemblance to Delayed Response, the classic test for dorsolateral prefrontal function in the rhesus monkey, and indeed performance on AB and Delayed Response in the same animals in the present study was fully comparable. These findings provide direct evidence that AB performance depends upon dorsolateral prefrontal cortex in rhesus monkeys and indicates that maturation of dorsolateral prefrontal cortex may underlie the developmental improvement in AB performance of human infants from 7.5-12 months of age. This improvement marks the development of the ability to hold a goal in mind in the absence of external cues, and to use that remembered goal to guide behavior despite the pull of previous reinforcement to act otherwise. This confers flexibility and freedom to choose and control what one does.

Corticopontine projection patterns were studied after injections of an 3H-leucine and 3H-proline mixture into each of four distinct cortical fields within the inferior parietal lobule and dorsal prelunate gyrus. Different preferential patterns of pontine labeling were observed for each of the four cortical areas studied. Multiple injections across the dorsal aspect of the prelunate gyrus (area DP) yielded scattered patches of label limited to the dorsolateral pontine nuclear region. A single injection within the lateral intraparietal area (area LIP), located in the caudal portion of the lateral bank of the intraparietal sulcus resulted in a series of labeled patches across the dorsal tier of cells stretching across the dorsal portions of the dorsolateral, peduncular and dorsal pontine nuclei. Injection of the cortex on the caudal aspect of the inferior parietal convexity (area 7a) produced multiple patches of label along the lateral margin of the ventral, lateral, and dorsolateral nuclei. Injection of area 7b resulted in label along the lateral aspects of the ventral, lateral and dorsolateral nuclei, as seen with area 7a injections, as well as additional label in the ventromedial portions of the ventral, peduncular and paramedian pontine nuclei. These results provide supporting anatomic evidence for the functional subdivision of the inferior parietal lobule and dorsal aspect of the prelunate gyrus and provide new information about the organization of cortical projections to the primate pontine nuclei.

The present study sought to determine whether differential effects could be found in the free behavior of early-operated monkeys with selective removal of the dorsolateral and orbital prefrontal cortex. The early-operated monkeys were first observed in their home cages in stable living groups which had existed for at least 6 1/2 months (together condition) and then again immediately after they had been separated from one another (separated condition). The major results indicated that the monkeys with orbital lesions differed from unoperated controls in more ways than did the dorsolateral monkeys: in the together condition, the orbital group slept more and were more sedentary even when awake; in the separated condition, by contrast, they became hyperactive and spent most of their time in locomotion. These findings lend support to the notion, derived from studies of cognitive behavior in infant-operated monkeys, that functions of the orbital cortex, unlike those of the dorsolateral cortex, are not spared following brain damage in infancy. Further, the nature of the behavior exhibited by the impaired monkeys led to the suggestion that the orbital cortex may play an important role in modulating arousal mechanisms in the infant and adult monkey alike.