The visual system of humans and other primates has the remarkable ability to recognize objects despite tremendous variation in their appearance, due to changes in size, position, background, and viewpoint. While this ability is central to human visual perception, the underlying brain mechanisms are poorly understood, and transformation-tolerant recognition remains a major challenge in the development of artificial vision systems. Arguably, this is a consequence of the formidable complexity of the primate visual system and the relatively narrow range of experimental approaches that human and nonhuman primate studies allow. Although, traditionally, the invasive study of the neuronal basis of object vision has been restricted to non-human primate experiments, recently, rodents are merging as powerful models to study visual processing. However, successful use of rodents as models for studying visual object recognition crucially depends on the ability of their visual system to construct representations of visual objects that tolerate (i.e., remain relatively unchanged with respect to) the tremendous changes in object appearance produced, for instance, by size and viewpoint variation. As the first part of this Thesis, I addressed this question by training rats to categorize a continuum of morph objects resulting from blending two object prototypes. The resulting psychometric curve (reporting the proportion of responses to one prototype along the morph line) served as a reference when, in a second phase of the experiment, either prototype was briefly presented as a prime, immediately before a test morph object. The resulting shift of the psychometric curve showed that recognition became biased (primed) toward the identity of the prime. Critically, this bias was observed also when the primes were transformed along a variety of dimensions (i.e., size, position, viewpoint, and their combination) that the animals had never experienced before. These results indicate that rats spontaneously perceive different views/appearances of an object as similar (i.e., as instances of the same object) and argue for the existence of neuronal substrates underlying formation of transformation-tolerant object representations in rats. As the next step, I tried to characterize such neuronal substrates by performing multi- electrode neuronal recordings (in anesthetized rats exposed to a battery of visual objects) from five different cortical areas of the rat brain: primary visual cortex (V1) and four extrastriate areas (named LM, AL, LI and LL) that are located laterally to V1 and have been proposed as candidate stages of a putative rat visual shape processing stream,homologous to the monkey 5 ventral visual stream (Apart from area AL that probably belongs to dorsal pathway in rat). An object set consisting of 10 different objects, each transformed across a variety of axes (position, size, in-depth azimuth rotation and in-plane rotation) was used. I found that along the processing hierarchy V1->LM->LI->LL, receptive fields become progressively bigger, as well as the latency of the response. Using information theory I found that, as the information travels through this hierarchy, the fractional information that each cell carries about the luminance gradually decreases, whereas the fractional information about shape gradually increases. Accordingly, I found that neurons along this pathway become increasingly tolerant to transformations. This indicates that neurons along this hierarchy become progressively tuned to more complex visual attributes and become more tolerant to transformations, thus suggesting that the pathway V1->LM->LI->LL could be homologous to the primate ventral stream. Overall, the combination of this behavioral and neurophysiological studies will provide an unprecedented understanding of high-level visual processing in a rodent species.
|Titolo:||Behavioral and Neuronal Substrates of Invariant Object Recognition in Rats|
|Data di pubblicazione:||28-apr-2014|
|Appare nelle tipologie:||8.1 PhD thesis|