Institut Jean Nicod

Accueil > Séminaires/Colloques > Archives > Séminaires > 2013-2014 > Perception > Présentation





L'équipe Perception, co-dirigée par Roberto Casati et Jérôme Dokic, organise un certain nombre de séminaires sur des thèmes à l’interface entre recherches conceptuelles et recherches empiriques sur la perception.
Ces séances seront l’occasion de connaître les travaux en cours des membres de l’équipe ainsi que de rencontrer des expert(e)s invité(e)s. Les axes principaux de recherche de l’équipe sont la nature de la perception, les sentiments perceptifs, et la perception des/par les artefacts cognitifs.

Contact : Valeria Giardino

Institut Jean-Nicod, Pavillon Jardin, 29, rue d'Ulm 75005 Paris. Salle de réunion, RDC.



14 Janvier 2014,  de 11h à 13h
Jessica Hartcher (IJN)

"Experiencing space: exteriorisation through touch"

Résumé: In this presentation I look at the process of distal attribution through sensory substitution. Sensory substitution allows us to measure the emergence of distalisation as it occurs online during device use. I propose an objective measure and discuss the assumptions involved. In three studies I further explore the issues of mapping tactile information into visual space and discuss the implications for our representation of an external world.

17 février de 10h à 12h, Salle de réunion du Pavillon Jardin
Gabriel Greenberg (UCLA),
"Reference and Predication in Pictorial Representation"

Résumé :
How do pictorial images express their content? Linguistic content arises, in the simplest case, when a sentence expresses the predication of a property to a referent. In the first part of this talk, I defend the familiar view that pictorial content also involves the attribution of properties to referents. Thus a given picture might depict Obama (the referent) as lifting his arm (the property). Unfortunately for this otherwise plausible account, the format of pictorial representation resists any neat structural division into subject and predicate, in the manner of language or logic.  How then is predicative pictorial content possible? In the second part of the talk, I extract a “baseline” semantics for iconic representations that answers this challenge, drawn from various recent accounts of pictures, maps, and diagrams. Put simply: such images do not mark a structural distinction between subject and predicate. Instead, the very same  features both refer to individual objects and express the properties ascribed to these objects.  In the final part of the talk, I show how this baseline semantics must be extended to account for the essential role of perspective or viewpoint in pictorial representation.

27 février de 11h à 13h, Salle de réunion du Pavillon Jardin
François Le Corre (IJN),
"Cross-linguistic support for the universality of the belief in five senses"

Sensory Categorization Relativism (SCR) is the thesis according to which the traditional Westerner conception of the senses is relative (Howes, 1991, 2003, Classen, 1993b, 1997). In particular, some people have recently argued that the enumeration of the senses in five or our belief in five senses is a cultural variant (Cassaniti & Luhrmann, 2011). In this paper, I argue that such thesis is unmotivated. In the first section, I introduce and discuss certain arguments in favour of SCR: the argument from sensory category ranking (Geurts, 2002) and the argument of cross-linguistic variations (Ritchie, 1991). In the second section, I focus on cross-linguistic variations and introduce a universalist interpretation of the lexical specialization of the verbs of sensory perception (Viberg, 1984, Majid & Levinson, 2007). In particular, I put forward a preliminary model of the classification of our sensory lexicon—à la Berlin & Kay (1969)—based upon the idea that our folk enumeration of the senses in five results (i) from the salience of our sense organs (eyes, ears, nose, skin and tongue) and (ii) from the spatial relation that those senses entertain with environmental information. Ultimately, I argue that this model could support the thesis that our belief in five senses is not conventional (in the Lewisian sense of the word), contrary to what Nudds (2004) claims.

17 avril de 11h à 13h
Joulia Smortchkova (IJN),
"Social Perception"

Résumé :
Can we see (some) emotions in another person’s face and body via perception or do we always cognitively infer their emotional state on the basis of low-level visual information about their expression and behavior? How should we understand “seeing emotions” in order for the perceptual hypothesis to come out (non-trivially) true? In my presentation I will argue that there are cases in which emotions can be perceived as such in the absence of concepts for these emotions. I will support the claim that we can-see some emotions by appealing to three sorts of arguments: a) the specific way in which the relevant perceptual states carry information about another’s emotion; b) the possibility of non-conceptual “seeing as”; c) psychological and neurological data on emotion perception. Taken singularly, these sources of evidence are not enough to conclusively establish the thesis of emotion perception, but taken together they are sufficient to make a solid case for there being at least some cases in which emotion is perceived rather than inferred on the basis of perception of low-level properties.

12 mai de 10h à 12h,
Martin Fortier (IJN)
"Disentangling noetic feelings from perceptual content in hallucinogenic experiences: A tentative proposal"

When asked about the reality status of the things seen under ayahuasca or psilocybin (two powerful hallucinogenic substances) psychonauts usually reply that during and after their experience they have no difficulty in distinguishing what belongs to the ordinary world and what belongs to the hallucinatory effects – and yet they refuse to gather from this that the things they hallucinate have no existence whatsoever. Hence the following conundrum: psychonauts report that hallucinatory entities they experience under hallucinogens can readily be discriminated from other entities they ordinarily encounter in the non-hallucinatory world, but they nonetheless refrain from saying that these hallucinatory entities are not real.
Non-experimental philosophers cannot help us solving this conundrum since the kind of conceptually-defined hallucinations they are discussing at length in their armchairs shares almost nothing with the experimental hallucinations we are examining here. Contrary to armchair philosophers, empirically-oriented ones offer us quite promising hints. In a very stimulating paper Dokic & Martin (2012) have suggested that two distinct components which are too often lumped together should be disentangled: on the one hand are sensorial contents of visual experience, and on the other are noetic or metacognitive feelings whose role is to tag sensorial contents as being familiar (or unfamiliar), difficult to process (or fluent), internally-generated (or externally-generated), real (or unreal), etc.
Dokic & Martin’s theory was first intended to shed light on hallucinations in which subjects are unable to discriminate hallucination from perception – from this point of view, their theory cannot properly help us solving our conundrum. However, it is my contention that mutatis mutandis the dual-phenomenology approach might be able to shed new light on hallucinogenic experiences. My conjecture is the following: psychonauts might suffer from a metacognitive dysfunction which leads them to tag (more precisely: to feel) hallucinatory entities as “hyper-real” – hence the persistent discriminability between hallucinatory entities (which are experienced as “hyper-real”) and ordinary entities (which are experienced as “real”) and hence the robust reluctance to assess the former as “unreal” while assessing the latter as “real”. Interestingly enough, first-person reports in which psychonauts say that hallucinatory entities are “more real than real” seem to corroborate the view I am tentatively putting forward.
After having presented the conundrum posed by hallucinogenic experiences and proposed the foregoing conjecture, I will explore mechanisms which could possibly explain such a metacognitive dysfunction and I will do so by comparing the phenomenology and neurophysiology of hallucinogenic experiences to that of the feeling of presence as reported by patients suffering from Parkinson’s disease, to that of the feeling of unreality as reported by patients suffering from derealisation and to that of the feeling of reality as experienced in virtual reality. Finally, I will outline the data collection project I am currently preparing and thanks to which I hope to adjudicate between two theories: one which claims, in the vein of orthodox theories of religious cognition (e.g., Boyer’s or Atran’s), that the supernatural character of entities stems from their minimal violation of intuitive laws; and another, in line with recent studies in “procedural metacognition” (Proust), which proposes that supernaturalness depends upon metacognitive feelings rather than observable content.

12 juin de 10h à 12h,
Brent Strickland (IJN),
"Does core cognition pose a problem for the modularity of vision?"

Résumé :
From as early as a few hours out of the womb, pre-verbal infants display a basic grasp of "naive physics", which includes representations of physical objects and of contact causality. For example, they understand that physical objects must follow spatially and temporally continuous paths (the continuity principle), that objects maintain single cohesive boundaries (the cohesion principle), and that upon contact, one object can cause another to move. In developmental psychology, this early emerging "knowledge" has traditionally been described in terms of infants' ability to conceptually reason about the physical world. The idea is that infants may have innate concepts that allow them to infer or deduce how objects will behave in the near future (in a way analogous to how an adult might infer, for example, that it will rain based on their seeing dark clouds). In contrast to the developmental findings, recent evidence in adults has emphasized the perceptual role of naive physics. This work has demonstrated that similar representations of objects and causality function to structure automatic processes of visual perception and attention. One potential way of explaining these adult findings is by claiming that early emerging physical concepts exert a top-down influence on the processes of visual perception. However, such an explanation has a major drawback in that it would be a counter example to the modularity of vision, which is the standard idea from vision science that perceptual processes are inherently divorced from higher-level cognition. Here I explore some theoretical options which are available to the defender of modularity, and argue in particular that  the infant results on core object knowledge could be re-interpreted as reflecting high-level, but still modularized, processes of the developing visual system.