IVOR · Neuronal substrates of invariant visual object recognition in rats

Neuronal substrates of invariant visual object recognition in rats

During the 4-year period of support of the Marie Curie IRG, we made considerable progress on both our original aims, that is: 1) to uncover the perceptual strategy underlying rat visual object recognition; and 2) to understand how visual object information is processed along rat occipitotemporal visual cortical areas. With regard to the first aim, we have discovered that rat object recognition relies on a combination of multiple visual features and that these features are mostly preserved across the transformations the objects undergo (these findings had been published in the Journal of Neuroscience [1]). In a follow-up study (published in the Frontiers on Neural Circuits [2]), we investigated the impact of stimulus discriminability on rat pattern vision, focusing on the difference between the perceptual strategies underlying the recognition of structurally similar vs. dissimilar objects. We found that the pattern of diagnostic features underlying the discrimination of highly similar objects were more scattered, more view-dependent, and more subject dependent in the case of similar objects, as compared to the case of dissimilar ones. These results suggest that in rats, as in primates, transformation-tolerant recognition can flexibly rely on either view-invariant representations of distinctive object features or view-specific representations that are acquired through exposure to multiple object views. With regard to the second aim, we made significant progress toward understanding the processing of visual object information across the succession of rat occipitotemporal visual areas. We performed multi-electrode neuronal recordings from four different areas belonging to this succession and we found that: 1) the more lateral (i.e., temporal) areas code higher-order visual features, as compared to the more medial ones; and 2) such a coding is more tolerant to variation in object appearance (e.g., position or size changes) in the more lateral areas, as compared to the more medial ones. Taken together with the progressive increase of receptive field size and response latency across the succession, these findings suggest that rat occipitotemporal visual areas may play the role of an object processing pathway, similar to the primate ventral stream. A manuscript describing these findings is currently under preparation. In addition to these research papers, the fellow has published two review articles about visual object recognition, respectively, in primates [3] and rats [4]. Overall, the research carried out with the support of the Marie Cure grant led to important advancements in our understanding of the processing of visual object information by the rat brain, thus fostering the use of such an animal model in vision studies. This research has also considerably helped the career development and reintegration of the fellow, who has been recently promoted to Associate Professor in his host institute for the project (i.e., SISSA). 1. Alemi-Neissi, A., Rosselli, F.B. & Zoccolan, D. J. Neurosci. 33, 5939–5956 (2013). 2. Rosselli, F.B., Alemi, A., Ansuini, A. & Zoccolan, D. Front. Neural Circuits 9, 10 (2015). 3. DiCarlo, J.J., Zoccolan, D. & Rust, N.C. Neuron 73, 415–434 (2012). 4. Zoccolan, D. Behav. Brain Res. 285, 10–33 (2015).

Data: CORDIS, © European Union