We study a self-organising neural network model of how visual representations in the primate dorsal visual pathway are transformed from an eye-centred to head-centred frame of reference. The model has previously been shown to robustly develop head-centred output neurons with a standard trace learning rule, but only under limited conditions. Specifically it fails when incorporating visual input neurons with monotonic gain modulation by eye-position. Since eye-centred neurons with monotonic gain modulation are so common in the dorsal visual pathway, it is an important challenge to show how efferent synaptic connections from these neurons may self-organise to produce head-centred responses in a subpopulation of postsynaptic neurons. We show for the first time how a variety of modified, yet still biologically plausible, versions of the standard trace learning rule enable the model to perform a coordinate transformation from eye-centred to head-centred reference frames when the visual input neurons have monotonic gain modulation by eye-position.

No abstract

We study a self-organising neural network model of how visual representations in the primate dorsal visual pathway are transformed from an eye-centred to head-centred frame of reference. The model has previously been shown to robustly develop head-centred output neurons with a standard trace learning rule [1], but only under limited conditions. Specifically it fails when incorporating visual input neurons with monotonic gain modulation by eye-position. Since eye-centred neurons with monotonic gain modulation are so common in the dorsal visual pathway, it is an important challenge to show how efferent synaptic connections from these neurons may self-organise to produce head-centred responses in a subpopulation of postsynaptic neurons. We show for the first time how a variety of modified, yet still biologically plausible, versions of the standard trace learning rule enable the model to perform a coordinate transformation from eye-centred to head-centred reference frames when the visual input neurons have monotonic gain modulation by eye-position. Author summaryCoordinate transformations are an essential aspect of behaviour. For instance, sensory information encoded in the coordinates of the retina needs to be transformed to relevant coordinates for planning and movement. Particularly, head-centred coordinates are essential for accurate motor behaviours and required to compute more complex coordinate transformations for sensorimotor integration [2]. Head-centred coordinates are obtained by combining information about the the retinal location of visual stimuli and the position of the eye. Previous work did not address the influence of different forms of gain modulation by eye position, albeit a variety of forms being widely reported for several cortical areas. Here we show how a biologically plausible model that successfully self-organised head-centred responses [1,3] fails when the visual input units have a commonly observed form of eye-position gain modulation, i.e. monotonic modulation. Our work makes an important contribution to understanding how head-centred responses may develop in the brain through an unsupervised process of visually-guided learning using a set of more sophisticated, and yet still biologically

Many researchers have tried to model how environmental knowledge is learned by the brain and used in the form of cognitive maps. However, previous work was limited in various important ways: there was little consensus on how these cognitive maps were formed and represented, the planning mechanism was inherently limited to performing relatively simple tasks, and there was little consideration of how these mechanisms would scale up. This paper makes several significant advances. Firstly, the planning mechanism used by the majority of previous work propagates a decaying signal through the network to create a gradient that points towards the goal. However, this decaying signal limited the scale and complexity of tasks that can be solved in this manner. Here we propose several ways in which a network can can self-organize a novel planning mechanism that does not require decaying activity. We also extend this model with a hierarchical planning mechanism: a layer of cells that identify frequently-used sequences of actions and reuse them to significantly increase the efficiency of planning. We speculate that our results may explain the apparent ability of humans and animals to perform model-based planning on both small and large scales without a noticeable loss of efficiency.

Historical data provide valuable information for the understanding of human interactions through time. However, mining this data is challenging as the available records are generally noise digitized handwritten, typewritten or press printed documents. In this research proposal, we plan to develop tools and techniques for pre-processing and extracting information from documents of the military dictatorship period that ruled Brazil from 1964 to 1985. The data to be analyzed consists of digitized images of records from DEOPS/SP (São Paulo State Department of Political and Social Order), an emblematic police agency which have monitored (and in some cases, harassed and tortured) hundreds of thousands Brazilian citizens during that period. The idea is to use state-of-the-art powerful artificial intelligence algorithms in conjunction with crowdsourcing techniques to preprocess and extract information from this important period of the Brazilian History."Those who cannot remember the past are doomed to repeat it" George Santayana