Neuronal activity was recorded from the superior colliculus (SC) and the underlying reticular formation in two monkeys during an arm reaching task. Of 744 neurons recorded, 389 (52%) clearly modulated their activity with arm movements. The temporal activity patterns of arm-movement-related neurons often had a time course similar to rectified electromyograms (EMGs) of particular muscles recorded from the shoulder, arm or trunk. These reach cells, as well as the muscles investigated, commonly exhibited mono- or biphasic (less frequently tri- or polyphasic) excitatory bursts of activity, which were related to the (pre-)movement period, the contact phase and/or the return movement. The vast majority of reach cells exhibited a consistent activity pattern from trial to trial as did most of the muscles of the shoulder, arm and trunk. Similarities between the activity patterns of the neurons and the muscles were sometimes very strong and were especially notable with the muscles of the shoulder girdle (e.g. trapezius descendens, supraspinatus, infraspinatus or the anterior and medial deltoids). This high degree of co-activation suggests a functional linkage, though not direct, between the collicular reach cells and these muscles. Neuronal activity onset was compared with that of 25 muscles of the arms, shoulders and trunk. The majority of cells (78.5%) started before movement onset with a mean lead time of 149+/-90 ms, and 36.5% were active even before the earliest EMG onset. The neurons exhibited the same high degree of correlation (r=0.97, Spearman rank) between activity onset and the beginning of the arm movement as did the muscles (r=0.98) involved in the task. The mean neuronal reach activity (background subtracted) ranged between 7 and 193 impulses/s (mean 40.5+/-24.2). The mean modulation index calculated [(reach activity background activity)/reach activity+background activity)] was 0.75+/-0.23 for neurons (n=358) and 0.87+/-0.14 for muscles (n=25). As the monkeys fixated the reach target constantly during an arm movement, neuronal activity which was modulated in this period was not related to eye movements. The three neck muscles investigated in the reach task exhibited no reach-related activity modulation comparable to that of either the reach cells or the muscles of the shoulder, arm and trunk. However, tonic neck muscle EMG was monotonically related to horizontal eye position. The clear skeletomotor discharge characteristics of arm-movement-related SC neurons revealed in this study agree with those already known from other sensorimotor regions (for example the primary motor, the premotor and parietal cortex, the basal ganglia or the cerebellum) and are consistent with the possible role of this population of reach cells in the control of arm movements.

Current views of the visual system assume that the primate brain analyses form and motion along largely independent pathways; they provide no insight into why form is sometimes interpreted as motion. In a series of psychophysical and electrophysiological experiments in humans and macaques, here we show that some form information is processed in the prototypical motion areas of the superior temporal sulcus (STS). First, we show that STS cells respond to dynamic Glass patterns, which contain no coherent motion but suggest a path of motion. Second, we show that when motion signals conflict with form signals suggesting a different path of motion, both humans and monkeys perceive motion in a compromised direction. This compromise also has a correlate in the responses of STS cells, which alter their direction preferences in the presence of conflicting implied motion information. We conclude that cells in the prototypical motion areas in the dorsal visual cortex process form that implies motion. Estimating motion by combining motion cues with form cues may be a strategy to deal with the complexities of motion perception in our natural environment.

We recorded from 389 "reach" neurons (two monkeys) in the superior colliculus (SC) and underlying reticular formation (RF) or adjacent periaqueductal grey, whose activity was related to visually guided arm movements. Reach neurons were present from approximately 0.7 mm down to a depth of 6 mm below the surface of the SC (mean 3.7+/-1.3, n=389). Although this mean distribution was different from that of cells with visual (mean depth 1.7+/-1.4 mm, n=283) or saccadic responses (mean depth 2.0+/-1.4 mm, n=232), there was a large amount of overlap. Fifty-five per cent of all reach cells (213/389) were assumed to be located inside the SC. The others were considered to be located in the underlying RF. The characteristics of visual responses and saccadic bursts (e.g. response latencies, discharge rates, burst durations) of arm-movement-related neurons were not different from those of typical visual or saccade cells in the SC. Although reach neurons could be recorded in a large area of the SC, they were found more often in the lateral than in the medial parts (chi-squared=19.3, P<0.001). Possible pathways by which arm-movement-related neuronal activity in and below the SC might gain access to spinal motor structures are discussed. The location of arm-movement-related neurons described in this study is in accordance with the known target areas of skeletomotor-related corticotectal projections and with the sites of origin of tectofugal pathways. It is concluded that this population of reach cells is in a position to relay and transmit limb movement information to the spinal motor system, where it may influence (or interact with) motor commands coming from other motor centres.

There is growing evidence that in primate cerebral cortex the areas along the 'dorsal pathway' are involved in the transformation of visual motion information towards a motor command. To pursue this cortical flow of information from visual motion areas to the motor cortex, single-cell activity was recorded from visual areas MT/MST (middle temporal area/medial superior temporal area) and from primary motor cortex (M1) while monkeys tracked moving targets with their right hand. Spike activity of 353 directionally tuned motor cortex cells was combined to a time-varying population vector, and similarly a time-resolved visual population vector was calculated from 252 MT/MST cells. Both population vectors code faithfully for the direction of the collinear motion of target and hand. For a given direction, the length of the population vectors varied over time during the performance of the task. The temporal evolution of both population responses reflects the different relationship between the early visual responses to the moving target and the directional motor command controlling the hand movement. The results indicate that during the visual tracking task visual and motor populations which code for similar directions of movement are co-activated with considerable temporal overlap. Despite this co-activation in both modalities, we failed to observe any significant synchronization between areas MT/MST and M1.