Thalamic Neurons in the Pigeon Compute Distance-to-Collision of an Approaching Surface

Abstract

The thalamofugal and tectofugal pathways in birds are two parallel visual pathways to the telencephalon and might be comparable to the geniculocortical and colliculo-pulvinar-cortical pathways in mammals, respectively. It is known that some tectal neurons in the tectofugal pathway can signal the time-to-collision of an approaching object. Here we show by single cell recording in the pigeon that a population of visual neurons in the nucleus opticus principalis thalami (nOPT) in the thalamofugal pathway is able to detect the distance-to-collision of a large surface approaching towards the animal. These neurons began response firing when the surface reached a threshold distance to the viewing eye and thereafter their firing rates increased exponentially until collision occurred at distance = zero. The response onset distance is nearly constant for a wide range of stimulus velocities and is equal to the product of velocity of approaching stimulus and response onset time of a nOPT cell. Furthermore, onset distance of looming responses in nOPT cells is close to that causing cardioacceleration in the pigeon viewing the approaching surface. It appears that nOPT and tectal neurons are dichotomized functionally to a large extent for detecting imminent dangers, and the pigeon’s ability to compute the distance-to-collision of an approaching surface may benefit avoiding large obstacles during flight.

Neuronal Responses to Looming Objects in the Superior Colliculus of the Cat

“…In subsequent experiments, the excitatory receptive field (ERF) was mapped for some looming-sensitive neurons. A black square of 4° (visual angle) was moved at the optimal velocity, randomly along a series of parallel paths covering the whole visual space [Wu et al, 2005;Xiao et al, 2006;Liu et al, 2008]. A contour map was obtained by thin-plate spline interpolation to illustrate the spatial distribution of evoked responses, and the ERF size was determined as the geometric mean of length and width of the area within which the moving square elicited a significant increase in the firing rate of the cell.…”

“…However, the presence of suppressive surround, usually several times larger than the excitatory region, was visible in many maps. When necessary, the extent of suppressive surround was estimated by gradually lengthening the square into a bar in the dimension perpendicular to that of motion until the firing rate evoked by the moving bar was reduced to a minimum [Wu et al, 2005;Liu et al, 2008].…”

“…Conceivably, the visual system plays an important role in processing the characteristic information of looming objects and producing the neural signals to trigger behavioral responses [van der Weel and van der Meer, 2009;Billington et al, 2011]. In fact, looming-sensitive visual neurons have been recorded in the moth [Wicklein and Strausfeld, 2000], the mantis [Yamawaki and Toh, 2009], the locust Simmons, 1992, 1997;Gray et al, 2010;Rogers et al, 2010], the crab , fish [Gallagher and Northmore, 2006], the frog [Nakagawa and Hongjian, 2010] and the pigeon [Wang and Frost, 1992;Liu et al, 2008]. Further studies have characterized several specific movement detector neurons in the locust optic lobe [Hatsopoulos et al, 1995;Gabbiani et al, 1999Gabbiani et al, , 2001Gabbiani et al, , 2004 and three classes of visual neurons in the pigeon tectofugal pathway [optic tectum : Wu et al, 2005; nucleus rotundus: Sun and Frost, 1998; telencephalic entopallium: Xiao et al, 2006] by their distinct response patterns to looming stimuli, and it has been suggested that these cells might be involved in computation of different optical variables of approaching objects.…”

Haptic distal spatial perception mediated by strings: Haptic “looming”.

“…This conjoint understanding of tau-like relationships seems likely to both help understand past results and to suggest future research opportunities. Liu, Niu, and Wang (2008) have found neurophysiological evidence in pigeons that appears to support this conception, although their data suggest a neural separation of function.…”

Visual discrimination thresholds for time to arrival

“…This entails that a "cognitive" strategy to compute t A by dividing perceived distance and perceived velocity is unlikely to be successful-if it is applied at all (Tresilian, 1991). Still, observers may attempt to infer distances from apparent object sizes (Gibson, 1950) and respond to distance to arrival (d A ) instead of t A (see Liu, Niu, & Wang, 2008, for evidence of accordingly tuned neurons in the pigeon, although d A is believed to be obtained from velocity-and τ G -sensitive neurons, on the basis of the equality of d = v·t). In a 2AFC scenario, such a response strategy would show up as overreliance on the difference of visual angles (Δθ i,j )-exactly what we found.…”