From the article: Abstract Over the last decades, philosophers and cognitive scientists have argued that the brain constitutes only one of several contributing factors to cognition, the other factors being the body and the world. This position we refer to as Embodied Embedded Cognition (EEC). The main purpose of this paper is to consider what EEC implies for the task interpretation of the control system. We argue that the traditional view of the control system as involved in planning and decision making based on beliefs about the world runs into the problem of computational intractability. EEC views the control system as relying heavily on the naturally evolved fit between organism and environment. A ‘lazy’ control structure could be ‘ignorantly successful’ in a ‘user friendly’ world, by facilitating the transitory creation of a flexible and integrated set of behavioral layers that are constitutive of ongoing behavior. We close by discussing the types of questions this could imply for empirical research in cognitive neuroscience and robotics.
LINK
Over the past decade, journalists have created in-depth interactive narratives to provide an alternative to the relentless 24-hour news cycle. Combining different media forms, such as text, audio, video, and data visualisation with the interactive possibilities of digital media, these narratives involve users in the narrative in new ways. In journalism studies, the convergence of different media forms in this manner has gained significant attention. However, interactivity as part of this form has been left underappreciated. In this study, we scrutinise how navigational structure, expressed as navigational cues, shapes user agency in their individual explorations of the narrative. By approaching interactive narratives as story spaces with unique interactive architectures, in this article, we reconstruct the architecture of five Dutch interactive narratives using the walkthrough method. We find that the extensiveness of the interactive architectures can be described on a continuum between closed and open navigational structures that predetermine and thus shape users’ trajectories in diverse ways.
DOCUMENT
We present a novel architecture for an AI system that allows a priori knowledge to combine with deep learning. In traditional neural networks, all available data is pooled at the input layer. Our alternative neural network is constructed so that partial representations (invariants) are learned in the intermediate layers, which can then be combined with a priori knowledge or with other predictive analyses of the same data. This leads to smaller training datasets due to more efficient learning. In addition, because this architecture allows inclusion of a priori knowledge and interpretable predictive models, the interpretability of the entire system increases while the data can still be used in a black box neural network. Our system makes use of networks of neurons rather than single neurons to enable the representation of approximations (invariants) of the output.
LINK
