We describe the operation of a midbrain neural circuit in pigeons that may participate in selecting and attending to one visual stimulus from the myriad displayed in their visual environment. This mechanism is based on a topographically organized cholinergic signal reentering the optic tectum (TeO). We have shown previously that, whenever a visual stimulus activates neurons in a given tectal location, this location receives a strong bursting feedback from cholinergic neurons of the nucleus isthmi pars parvocellularis (Ipc), situated underneath the tectum. Here we show that, if a second visual stimulus is presented, even far from the first, the feedback signal to the first tectal location is diminished or suppressed, and feedback to the second tectal location is initiated. We found that this long-range suppressive interaction is mostly mediated by the nucleus isthmi pars magnocellularis, which sends a wide-field GABAergic projection to Ipc and TeO. In addition, two sets of findings indicate that the feedback from the Ipc modulates the ascending output from the TeO. First, visually evoked extracellular responses recorded in the dorsal anterior subdivision of the thalamic nucleus rotundus (RtDa), receiving the ascending tectal output, are closely synchronized to this feedback signal. Second, local inactivation of the Ipc prevents visual responses in RtDa to visual targets moving in the corresponding region of visual space. These results suggest that the ascending transmission of visual activity through the tectofugal pathway is gated by this cholinergic re-entrant signal, whose location within the tectal visual map is dynamically defined by competitive interactions.

The retinotectofugal system is the main visual pathway projecting upon the telencephalon in birds and many other nonmammalian vertebrates. The ascending tectal projection arises exclusively from cells located in layer 13 of the optic tectum and is directed bilaterally toward the thalamic nucleus rotundus. Although previous studies provided evidence that different types of tectal layer 13 cells project to different subdivisions in Rt, apparently without maintaining a retinotopic organization, the detailed spatial organization of this projection remains obscure. We reexamined the pigeon tectorotundal projection using conventional tracing techniques plus a new method devised to perform small deep-brain microinjections of crystalline tracers. We found that discrete injections involving restricted zones within one subdivision retrogradely label a small fraction of layer 13 cells that are distributed throughout the layer, covering most of the tectal representation of the contralateral visual field. Double-tracer injections in one subdivision label distinct but intermingled sets of layer 13 neurons. These results, together with the tracing of tectal axonal terminal fields in the rotundus, lead us to propose a novel "interdigitating" topographic arrangement for the tectorotundal projection, in which intermingled sets of layer 13 cells, presumably of the same particular class and distributed in an organized fashion throughout the surface of the tectum, terminate in separate regions within one subdivision. This spatial organization has significant consequences for the understanding of the physiological and functional properties of the tectofugal pathway in birds.

Fast oscillatory bursts (OBs; 500 -600 Hz) are the most prominent response to visual stimulation in the optic tectum of birds. To investigate the neural mechanisms generating tectal OBs, we compared local recordings of OBs with simultaneous intracellular and extracellular single-unit recordings in the tectum of anesthetized pigeons. We found a specific population of units that responded with burst discharges that mirrored the burst pattern of OBs. Intracellular filling with biocytin of some of these bursting units demonstrated that they corresponded to the paintbrush axon terminals from the nucleus isthmi pars parvocellularis (Ipc). Direct recordings in the Ipc confirmed the high correlation between Ipc cell firing and tectal OBs. After injecting micro-drops of lidocaine in the Ipc, the OBs of the corresponding tectal locus disappeared completely. These results identify the paintbrush terminals as the neural elements generating tectal OBs. These terminals are presumably cholinergic and ramify across tectal layers in a columnar manner. Because the optic tectum and the Ipc are reciprocally connected such that each Ipc neuron sends a paintbrush axon to the part of the optic tectum from which its visual inputs come, tectal OBs represent re-entrant signals from the Ipc, and the spatial-temporal pattern of OBs across the tectum is the mirror representation of the spatial-temporal pattern of bursting neurons in the Ipc. We propose that an active location in the Ipc may act, via bursting paintbrushes in the tectum, as a focal "beam of attention" across tectal layers, enhancing the saliency of stimuli in the corresponding location in visual space.

When a salient object in the visual field captures attention, the neural representation of that object is enhanced at the expense of competing stimuli. How neural activity evoked by a salient stimulus evolves to take precedence over the neural activity evoked by other stimuli is a matter of intensive investigation. Here, we describe in pigeons (Columba livia) how retinal inputs to the optic tectum (TeO, superior colliculus in mammals), triggered by moving stimuli, are selectively relayed on to the rotundus (Rt, caudal pulvinar) in the thalamus, and to its pallial target, the entopallium (E, extrastriate cortex). We show that two satellite nuclei of the TeO, the nucleus isthmi parvocelullaris (Ipc) and isthmi semilunaris (SLu), send synchronized feedback signals across tectal layers. Preventing the feedback from Ipc but not from SLu to a tectal location suppresses visual responses to moving stimuli from the corresponding region of visual space in all Rt subdivisions. In addition, the bursting feedback from the Ipc imprints a bursting rhythm on the visual signals, such that the visual responses of the Rt and the E acquire a bursting modulation significantly synchronized to the feedback from Ipc. As the Ipc feedback signals are selected by competitive interactions, the visual responses within the receptive fields in the Rt tend to synchronize with the tectal location receiving the "winning" feedback from Ipc. We propose that this selective transmission of afferent activity combined with the cross-regional synchronization of the areas involved represents a bottom-up mechanism by which salient stimuli capture attention.

The sensory–motor division of the avian arcopallium receives parallel inputs from primary and high‐order pallial areas of sensory and vocal control pathways, and sends a prominent descending projection to ascending and premotor, subpallial stages of these pathways. While this organization is well established for the auditory and trigeminal systems, the arcopallial subdivision related to the tectofugal visual system and its descending projection to the optic tectum (TeO) has been less investigated. In this study, we charted the arcopallial area displaying tectofugal visual responses and by injecting neural tracers, we traced its connectional anatomy. We found visual motion‐sensitive responses in a central region of the dorsal (AD) and intermediate (AI) arcopallium, in between previously described auditory and trigeminal zones. Blocking the ascending tectofugal sensory output, canceled these visual responses in the arcopallium, verifying their tectofugal origin. Injecting PHA‐L into the visual, but not into the auditory AI, revealed a massive projection to tectal layer 13 and other tectal related areas, sparing auditory, and trigeminal ones. Conversely, CTB injections restricted to TeO retrogradely labeled neurons confined to the visual AI. These results show that the AI zone receiving tectofugal inputs sends top‐down modulations specifically directed to tectal targets, just like the auditory and trigeminal AI zones project back to their respective subpallial sensory and premotor areas, as found by previous studies. Therefore, the arcopallium seems to be organized in a parallel fashion, such that in spite of expected cross‐modal integration, the different sensory–motor loops run through separate subdivisions of this structure.