Archive Electronique - Institut Jean Nicod
Not a member yet
1997 research outputs found
Sort by
La véracité des news dans une perspective d'évolution culturelle
No full textNational audienceDe façon générale, les informations transmisses culturellement tendent à être justes si les personnes qui les transmettent bénéficient d’un bon feedback sur leur justesse, et à être fausses si cela n’est pas le cas. Cette règle semble bien s’appliquer à la transmission informelle de news : les rumeurs concernant assez directement les individus qui les transmettent tendent à être justes, alors que celles ne les concernant pas sont presque toujours fausses. Il en a longtemps été de même pour les news transmisses par la presse : jusqu’au 19ème siècle (au moins), la fiabilité de la presse était globalement assez basse. Dans ce cadre, les médias traditionnels (mainstream media) modernes sont une exception à la règle, puisque qu’ils transmettent des informations plutôt justes, alors même que leur audience n’est que très rarement en position de juger par eux même de la justesse de ces informations. Dans le cadre de l’évolution culturelle, je vais tenter d’expliquer les observations suivantes : (i) pourquoi les news tendent à être fausses en l’absence de bon feedback ; (ii) pourquoi un certain contexte institutionnel a rendu possible l’émergence de news généralement justes ; (iii) pourquoi l’émergence de ce type de news peut malheureusement avoir a des effets délétères sur l’environnement informationnel, tels qu’une certaine désaffection vis-à-vis des news, ou le retranchement de certains segments de la population dans le complotisme
Continuité et discontinuité entre les sciences cognitives et les sciences sociales
No full textNational audiencePlutôt qu’une présentation classique suivie d’une discussion, je voudrais engager un dialogue avec tous les participants sur les rapports entre sciences sociales et sciences cognitives. Pour cela, je présenterai quelques thèses qui seront vues, je l’espère, comme contestables et que nous pourrons discuter l’une après l’autre
Delusions as Epistemic Hypervigilance
No full textInternational audienceDelusions are distressing and disabling symptoms of various clinical disorders. Delusions are associated with an aberrant and apparently contradictory treatment of evidence, characterized by both excessive credulity (adopting unusual beliefs on minimal evidence) and excessive rigidity (holding steadfast to these beliefs in the face of strong counterevidence). Here we attempt to make sense of this contradiction by considering the literature on epistemic vigilance. Although there is little evolutionary advantage to scrutinizing the evidence our senses provide, it pays to be vigilant toward ostensive evidence—information communicated by others. This asymmetry is generally adaptive, but in deluded individuals the scales tip too far in the direction of the sensory and perceptual, producing an apparently paradoxical combination of credulity (with respect to one’s own perception) and skepticism (with respect to the testimony of others)
The art of the float
No full textInternational audienceFor more than 2000 years, artists have exploited cast shadows to influence how objects appear to be positioned in a scene. A contact cast shadow can anchor an object to the ground and a detached cast shadow can make an object appear to float. However, there is a period of approximately 1000 years when there were virtually no cast shadows in art. How were states of contact versus floating depicted by artists without cast shadows? Here, we survey various techniques used by artists to anchor relative position with and without cast shadows. We then conduct experimental tests of the hypothesized surface attraction principles that underlie these techniques. In the absence of cast shadows, an object (a wooden box) was often seen as resting on a surface as long as that surface offered information about ground orientation and support (a tiled floor). When the ground surface was ambiguous and cloud-like (1/f noise), the box was more likely to be seen to float. The presence of cast shadows made the box appear to contact the ground whether it was well-defined or ambiguous. Both shadows and surface support also increased the accuracy with which participants detected when the box was tilted up from the ground. These results indicate that artists long ago discovered the important power of support relationships to anchor objects to surfaces in the absence of shadows
Exhaustivity and anti-exhaustivity in the RSA framework: Testing the effect of prior beliefs.
No full textInternational audienceDuring communication, the interpretation of utterances is sensitive to a listener's probabilistic prior beliefs. In this paper, we focus on the influence of prior beliefs on so-called exhaustivity interpretations, whereby a sentence such as Mary came is understood to mean that only Mary came. Two theoretical origins for exhaustivity effects have been proposed in the previous literature. On the one hand are perspectives that view these inferences as the result of a purely pragmatic process (as in the classical Gricean view, and more recent Bayesian approaches); on the other hand are proposals that treat them as the result of an encapsulated semantic mechanism (Chierchia, Fox & Spector 2012). We gain traction on adjudicating between these two approaches with new theoretical and experimental evidence, focusing on the behavior of different models for exhaustivity effects, all of which fit under the Rational Speech Act modeling framework (RSA, Frank & Goodman, 2012). Some (but not all!) of these models include an encapsulated semantic mechanism. Theoretically, we demonstrate that many RSA models predict not only exhaustivity, but also anti-exhaustivity, whereby “Mary came” would convey that Mary and someone else came. We evaluate these models against data obtained in a new study which tested the effects of prior beliefs on both production and comprehension, improving on previous empirical work. We find that the models which have the best fit to human behavior include an encapsulated exhaustivity mechanism. We conclude that, on the one hand, in the division of labor between semantics and pragmatics, semantics plays a larger role than is often thought, but, on the other hand, the tradeoff between informativity and cost which characterizes all RSA models does play a central role for genuine pragmatic effects
Communicating AI intentions to boost Human AI cooperation
No full textInternational audienceInteracting with Artificial Intelligence (AI) profoundly changes the nature of human activity as well as the subjective experience that agents have of their own actions and their consequences. We propose to mitigate this effect by making AI systems more intelligible to human operators. We hypothesized that the readability of system intentions is a key element of their predictability and, by extension, of the human operator's abilities to interact effectively with highly automated systems. We conducted experiments to explore the impact of the communication of AI intentions during joint human-AI interaction (intention-based explanations). Trust human operators’ have towards such algorithms as well as their sense of control across different dimensions (performance, action fluency, contribution) was measured. Overall, our results suggest that adding intention-based explanations during human-AI interaction indeed support cooperation between the human operator and artificial agents
Plural and Quantified Protagonists in Free Indirect Discourse and Protagonist Projection
No full textInternational audienceIn this paper I observe a number of new plural and (apparently) quantified examples of free indirect discourse (FID) and protagonist projection (PP). I analyse them within major current theoretical approaches, proposing extensions to these approaches where needed. In order to derive the wide range of readings observed with plural protagonists, I show how we can exploit existing mechanisms for the interpretation of plural anaphora and plural predication. The upshot is that the interpretation of plural examples of perspective shift relies on a remarkable concert of covert semantic and pragmatic operations
Negative concord in Russian Sign Language
No full textInternational audienceIn this paper, we provide the first systematic description of negative concord in Russian Sign Language (RSL). Although non-manual markers have been argued to participate in negative concord in sign languages, negative concord involving only manual signs has been shown to be much rarer. The RSL pattern thus fills this typological gap, providing one of the first clear cases of negative concord of manual signs in sign language. We show that RSL contributes important new data to the typology of negative concord known from spoken language. First, we show that RSL (like Hungarian) shows characteristics of both strict and non-strict patterns of negative concord. In neutral contexts without movement, NC items in both subject and object position require a negative licensor; on the other hand, in contexts with appropriate information structure, focused NC items may appear in specific structural positions without a licensor. These facts provide evidence against analyzing the strict/non-strict divide as a language-level property. Second, focusing on non-strict concord, we show that RSL diverges from other languages with respect to important macro-syntactic properties. In RSL, like in a number of other sign languages, negative elements may appear in a structure on the right edge. It is precisely this position that allows NC items to appear without a licensor; in this respect, RSL is a mirror-image of languages like Italian. These syntactic properties provide new evidence that structural hierarchy, not linear order, is responsible for explaining the presence or absence of a licensor in patterns of non-strict concord
Why is violence high and persistent in deprived communities? A formal model
No full textInternational audienceThere is massive variation in rates of violence across time and space. These rates are positively associated with economic deprivation and inequality. They also tend to display a degree of local persistence, or ‘enduring neighbourhood effects’. Here, we identify a single mechanism that can produce all three observations. We formalize it in a mathematical model, which specifies how individual-level processes generate the population-level patterns. Our model assumes that agents try to keep their level of resources above a ‘desperation threshold’, to reflect the intuitive notion that one of people's priorities is to always meet their basic needs. As shown in previous work, being below the threshold makes risky actions, such as property crime, beneficial. We simulate populations with heterogeneous levels of resources. When deprivation or inequality is high, there are more desperate individuals, hence a higher risk of exploitation. It then becomes advantageous to use violence, to send a ‘toughness signal’ to exploiters. For intermediate levels of poverty, the system is bistable and we observe hysteresis : populations can be violent because they were deprived or unequal in the past, even after conditions improve. We discuss implications of our findings for policy and interventions aimed at reducing violence
Interpretability of statistical approaches in speech and language neuroscience
No full textInternational audienceThe classical view of the speech and language neural system is that of a hierarchy of interdependent modules, enabling the progressive transformation of a continuous acoustic stream into an articulated series of concepts. This modular and hierarchical view follows from the combination of lesion studies (double dissociations) and hypothesis-based factorial designs in which only a few sensory or cognitive factors are varied at a time. In the last ten years, however, data-driven explorations of large neuroimaging datasets have allowed for a more agnostic approach, and led to a whole new view where segregated hierarchically organized modules seem to give way to continuous multidimensional representations, with e.g. a distributed semantic system. While both approaches have brought about significant contributions to speech and language neuroscience, making coherent sense of them represents a substantial challenge. In this review article, we synthesize methodological and experimental findings from the speech and language neuroscience literature, dissecting strength and pitfalls of each approach and suggesting ways in which approaches could be integrated