1. The input-output connectivity, in cat, of tectoreticular (TRNs) and tectoreticulospinal (TRSNs) neurons [together called TR(S)Ns] suggests a role for these cells in the sensorimotor transformations necessary for controlling orienting behavior. Multimodal sensory information converges directly onto these tectal neurons, and they project to several brain stem and spinal cord centers involved in the control of eye- and head-orienting movements. In this and the following two papers, we describe the sensorimotor discharges of antidromically identified TR(S)Ns. Here we describe the process of localizing and identifying them, characteristics of both their antidromic and sensory responses, and effects of behavioral context on these responses. 2. In 13 alert, chronically prepared cats, a total of 293 neurons were antidromically identified from either the predorsal bundle (PDB) immediately rostral to abducens nucleus or the ventromedial funiculus of the spinal cord at the level of the first cervical vertebra (C1). The cell bodies of all identified TR(S)Ns were confined to the intermediate and deep laminae of the superior colliculus (SC). The antidromic nature of the action potential evoked by stimulating either the PDB or C1 was verified by the use of a number of established criteria, including collision testing. 3. The mean antidromic latency from the PDB (TRNs + TRSNs) was 0.84 +/- 0.59 (SD) ms (n = 217). The conduction velocities of all cells activated by PDB stimulation ranged from 4 to 40 m/s. The mean latency from C1 (TRSNs) was 1.03 +/- 0.52 ms (SD) (n = 64), whereas conduction velocities ranged from 14 to 80 m/s. 4. One hundred thirty-eight TR(S)Ns were studied long enough to yield significant data regarding their involvement in visuomotor-orienting behavior. Ninety-eight percent (130/133) of the TR(S)Ns tested for visual responses could be induced to discharge action potentials in response to some form of visual stimulation. The other three neurons remained silent, even in response to the most provocative stimuli. These silent neurons nevertheless were shown to be depolarized by visual stimuli. TR(S)Ns were occasionally tested for auditory and somatosensory responses and some were multimodal. 5. TR(S)Ns had visual receptive fields that conformed to the retinotopic map of the visual field that is represented within the SC. Cells found in the lateral SC had receptive fields located in the lower visual field, whereas neurons that were situated medially had receptive fields in the upper visual field. Cells found in the rostral SC had small fields that included a representation of the area centralis. TR(S)Ns located in the more caudal regions of the SC had large fields that did not include a central representation. To distinguish between rostral and caudal cell groups, and for reasons that will become amply evident in the subsequent papers, we call them fixation [fTR(S)Ns] and orientation [oTR(S)Ns], respectively. The present paper considers only the latter. 6. oTR(S)Ns responded to the onset of a stationary light-emitting diode (LED) with a brief phasic burst, the latency of which was remarkably constant from trial to trial. The mean onset latency for 33 cells was 57 ms. For this same population, the mean of the earliest response latency of each cell was 47 ms. 7. oTR(S)Ns responded very well to moving visual stimuli. All 15 neurons tested were found to have a directional preference for stimulus motion away from the area centralis. For most cells, the preferred direction of stimulus motion was parallel to the horizontal meridian. 8. During the behavioral act of attentive fixation, 10 of the 13 oTR(S)Ns that were tested for fixation effects had significantly attenuated responses to an abstract stimulus moving in their receptive field. 9. By comparison, the sensory responses of oTR(S)Ns were enhanced when the stimulus became the target of an orienting response. In 9 of the 14 cells tested, response to the onset of an LED was significantly augmented when the cat oriented to the LED within 1 s of its onset. The amount of enhancement that occurred was linked to the latency of the animal's orienting response. Enhancement was greatest when the animal's response latency was very short, such that the phasic visual response overlapped movement onset. 10. The excitability levels of 15 oTR(S)Ns were evaluated in different behavioral conditions, using a twin-pulse antidromic technique to determine whether the behaviorally induced variability in oTR(S)N sensory responsiveness was the result of changes in either the afferent sensory input or the level of oTR(S)N excitability. oTR(S)Ns were found to be at significantly lower levels of excitability when the hungry animal attentively fixated the food target. When the animal was about to orient to a food target moving through a cell's visual receptive field, the neuron was at a significantly higher level of excitability. If the animal was about to orient to the food in the opposite direction, the level of excitability was significantly reduced. Because oTR(S)Ns project directly onto eye and head premotor circuits, the sensory responsiveness of these cells is presumed to influence the probability of occurrence of orienting movements.
