Proceedings of the ESPP-2002 Congress
From 10th to 13th July 2002,
Institute of cognitive science
CNRS – Université Lyon 1

accomodation

access

The aim of the Society is "to promote interaction between philosophers and psychologists on issues of common concern". Psychologists, neuroscientists, linguists, computer scientists and biologists are encouraged to report experimental, theoretical and clinical work that they judge to have philosophical significance; and philosophers are encouraged to engage with the fundamental issues addressed by and arising out of such work. In recent years ESPP sessions have covered such topics as spatial concepts, simulation theory, attention, problems of consciousness, emotion, perception, early numerical cognition, infants' understanding of intentionality, memory and time, motor imagery, counterfactuals, the semantics/pragmatics distinction, reasoning, vagueness, mental causation, action and agency, thought without language, externalism, connectionism, and the interpretation of neuropsychological results.

Congress organization

Committee

Pierre Jacob, President Prof. Philosophy, Paris Naomi Eilan, Secretary Dir. Brit. Acad. Proj. Philosophy, Warwick Jérôme Dokic, Treasurer Philosophy, Institut Jean-Nicod, Paris Josef Perner, FBA Prof. Psychology, Salzburg James Hampton, Prof. Psychology, City Univ., London Beate Sodian, Prof. Psychology, München Albert Newen, Philosophy, Bonn Josep Macia, Philosophy, Barcelona  Ingar Brinck, Philosophy, Lund Tim Dalgleish, Psychology, MRC-CBU, Cambridge

Programme Chairs

Michael Martin, Philosophy, Univ. College, London Anton Kühberger, Psychology, Salzburg Bernhard Schröder, Communic. Res & Phonetics, Bonn

Advisory Board

Lawrence Weiskrantz, FRS Prof.em. Psychology, Oxford Christopher Peacocke, FBA Prof. Philosophy, Oxford Anthony Marcel, Sen. Scientist, MRC-CBU - Cambridge Hans Kamp, FBA Prof. Linguistics, Stuttgart Gerd Gigerenzer, Dir. MPI, Berlin Manuel Garcia-Carpintero, Prof. Philosophy, Barcelona Beatrice de Gelder, Prof. Psychology, Tilburg/Louvain Martin Davies, Prof. Philosophy, RSSS, Austral. Nat. Univ. John Campbell, Prof. Philosophy, Oxford Daniel Andler, Prof. Philosophy, Sorbonne/ENS-Paris

Local organization

Frédérique de Vignemont, Philosophy, Lyon Nausicaa Pouscoulous, Psycholinguistics, Lyon

Program – Congress ESPP 2002
 

Wednesday 10th July

8.00 – 8.45  Registration

8.45 – 9.00  Welcome

9.00 – 10.15   Plenary session
 Manuel Garcia Carpintero, Qualia that It Is Right to Quine

10.15 – 10.45 Coffee Break

10.45 – 1.00  Symposium
Self-recognition
 Convenor : Marc Jeannerod
 Speakers : Shaun Gallagher, Günther Knoblich, Thomas Metzinger

1.00 – 2.15  Lunch

2.15 – 4.30  Symposium
   Motor intention and the control of action
 Convenors: Y. Rossetti and E. Pacherie
 Speakers: H. Bekkering, E. Pacherie, L. Pisella, I. Toni

4.30 – 5.00  Coffee Break

5.00 – 6.00  Parallel Sessions
MIND-BODY PROBLEM
S. Walter, Multiple Realizability and Property-Identities: An Incompatible Couple?
H. Looren de Jong, Reduction, elimination and the dynamics of explanation
Luc Faucher & Pierre Poirier, The problem of the disunity of psychology and a possible answer to it
INTENTIONALITY AND BELIEFS
P. Raatikainen, Thinking what’s not there
S. Butterfill, Understanding belief
Manuel de Pinedo Garcia, Computation and Mind: From Syntax to Pragmatics
SOCIAL PSYCHOLOGY
D. Shmotkin, The Pursuit of Happiness in a Hostile World: The Role of Subjective Well-Being Reconsidered
A. Wallin; R. McElreath; M. Larsson, The epistemic nature of conformity
 

Thursday 11th July

9.00 – 10.15   Plenary session
Tim Dalgleish, Emotions and inhibitory control

10.15 – 10.45 Coffee Break

10.45 – 1.00  Symposium
Pragmatics
Convenor : Richard Breheny
Speakers : Robyn Carston, Bart Geurts, Charles Travis

1.00 – 2.30  Lunch

2.30 – 4.00   Parallel Sessions
CONCEPT
J. Tanney, Conceptual Analysis, Theory Construction, and Conceptual Elucidation
E. Machery, Should really philosophers rule over psychologists?
P. Chuard, From Concepts to Appearances
PERCEPTION
O. Vilarroya, A Sensational Crisis
D. Taraborelli, Feature Binding and Object Perception.
A. Tüscher, The seeing-in theory of depiction and the psychophysics of picture perception
LANGUAGE AND COMMUNICATION
N. Allott; P. Rubio, This paper fills a much-needed gap
N. Pouscoulous; I. Noveck; G. Politzer, Evidence for Scalar Implicatures in Young Children
J. Raczaszek –Leonardi, Symbolic and dynamic aspects of language

4.00 – 4.30 Coffee Break

4.30 – 6.00 Parallel Sessions
CONSCIOUSNESS
B. Kotchoubey; A. Kuebler; N. Birbaumer, Conscious perception of one's own brain processes
E. Bacon; L. Paire-Ficout; M. Izaute, Dissociation between subjective experience and the cognitive process under the effect of lorazepam : the case of the TOT state
J. Broackes, Colour Blindness, Colour Language and the Dimensions of Visual Appearance
THE SELF
A. Newstead, Agency, Perception, and Self-Awareness
D. Legrand, Me, I, Myself
T. Vierkant, Is There A Self? Two Crucial Distinctions
CATEGORY
R. Dietz, Is there such a thing as a vagueness-related conceptual role?
J. Hampton; I. Kempff-Andersen, Category-based Induction  - an effect of conclusion typicality
P. Rubio, Understanding nominal metaphors: beyond class inclusion

6.30 – 7.30  Commitee meeting
 
 

Friday 12th July

9.00 – 10.15  Plenary session
Matthew W. Crocker, Expectations in Incremental Language Comprehension

10.15 – 10.45 Coffee Break

10.45 – 1.00 Symposium
Game Theory and Language
Convenor : Robert van Rooy
Speakers : Arthur Merin, Nicholas Asher, Prashant Parikh

1.00 – 2.30  Lunch

2.30 – 4.00   Poster Session

4.00 – 4.30  Coffee Break

4.30 – 6.00  Parallel Sessions
LOGIC AND REASONING
M. Siebel, There’s something about Linda. Probability, coherence and rationality
J. Smallwood, Effort as an affordance : the consequences for our understanding of thinking
P. Nakov, Latent Semantic Analysis for Notional Structures Investigation
MENTAL CAUSATION
M. Schouten; F. Keijzer, Embedded Cognition and Wide Mental Causation
A. Newen ; R. Cuplinskas, Mental Causation: A real phenomenon in a physicalistic world without epiphenomenalism or overdetermination
A. Reboul, A reductio ad absurdum argument against philosophical zombies
THEORIES OF REFERENCE
E. Ahlstrom, Semantics-pragmatics interface: the referential-attributive distinction revisited
G. Marti, Substitution and Identity
A. Raftopoulos; V. C. Mueller, A Step Toward Solving the Grounding Problem

7.30   Conference dinner at the restaurant “Les muses de l’Opéra”

Saturday 13th July

9.00 – 10.15  Plenary session
Peter Gärdenfors, Concept learning in geometrical spaces

10.15 – 10.45 Coffee Break

10.45 – 1.00  Symposium
Hypnosis
Convenor: Zoltan Dienes
Speakers : Sakari Kallio, Paloma Enríque

1.00 – 2.30  Lunch

2.30 – 4.00   Parallel Sessions
LANGUAGE : acquisition and impairment
A. Oliveira, Psycholinguistics in second language learning
P. Bongioanni; G. Buoiano ; M. Magoni, Linguistic impairments in amyotrophic lateral sclerosis/motor disease
S. Kasvinov, Periodic Table of Children’s psyche development stages
THOUGHT AND LANGUAGE
A. Vicente; F. Martínez-Manrique, Linguistic Relativity and the Language of Thought
D. Kelly, Mental translations
M. Textor, How One Should Become A Sensible Millian Heir
ATTENTION AND IMAGINATION
I. Brinck, The pragmatics of imperative and declarative pointing
M. Doherty; M. Sprung; N. McGuigan, Here’s looking at you kid: Children’s understanding of eye-direction
S. Dorsch, The unity of imagining
 
 

 PLENARY SESSIONS

Matthew W. Crocker
Chair of Psycholinguistics, Saarland University (Germany)
Expectations in Incremental Language Comprehension: The Role of Context and Experience

The experimental psycholinguistic literature abounds with evidence that human language comprehension takes place incrementally: people construct rich, compositional interpretations, word-by-word, as utterances are heard or read. One important consequence of incrementality is that people are forced to resolve local ambiguities – occurring during an utterance – in the face of incomplete information. Furthermore, Tanenhaus et al. (1995) have shown that when sentences are heard simultaneously with presentation of a related visual scene, eye-movements can reveal much about what interpretation people adopt and how visual information is used.
Recent evidence, further suggests that the human language processor goes beyond incrementality, and is in fact anticipatory. That is, upon hearing the initial words of an utterance, the human language processor appears to construct hypotheses about what is likely to come next. Altmann & Kamide (1999), for example, followed subjects’ eye-movements in a scene containing a boy, a cake, and a number of toys, when accompanied by an unambiguous spoken sentences such as “The boy will eat the cake”. Upon hearing the verb “eat”, subjects launch anticipatory eye-movements to relevant objects (in this case, the cake) before those objects are heard.
In this talk I will present related studies for a range of ambiguous sentence constructions. These studies are novel in that they exploit evidence from anticipatory eye-movements to determine preferred interpretations when hearers encounter a temporary ambiguity. More specifically, the studies focus on ascertaining the influence of syntactic (Case marking) and probabilistic (word order) linguistic constraints, lexical (verb) information, and the use of plausibility and thematic role information available in a visual scene. The experiments focus on the Subject Verb Object (SVO) and Object Verb Subject (OVS) word order possibilities in German. While Subjects and Objects can be unambiguously Case marked (masculine), the feminine and neuter forms are ambiguous. In brief, I will summarise recent evidence emerging from our laboratory, for the following behaviours:
1. Rapid combination of Case information and verb argument structure to predict forthcoming arguments (Kamide et al, 2002). That is (unambiguous) Subject + Verb triggers anticipation of the Object, while Object + Verb triggers anticipation of the Subject.
2. A preference to interpret a sentence initial Case ambiguous NP as Subject/Nominiativ (rather then Accusative) and also as Agent (rather than Patient).
3. The rapid use of verb + scene information to resolve the SVO vs OVS ambiguity when the initial NP is case ambiguous.
4. The sensitivity of the Subject first preference to syntactic priming: the likelihood of interpreting an ambiguous sentence initial NP as Object is increased if listeners have just read an unambiguous OVS priming sentence.
Finally, I will discuss the implications of these results for computational models of human language comprehension. While several models of incremental sentence processing have been developed, not all are able to naturally account for these anticipatory hypotheses. I will sketch a solution which builds upon several experience-based models which have been proposed (Jurafsky, 1996; Crocker & Brants, 2000), arguing that anticipations must be modeled as expectations which are based on our prior linguistic experience (Crocker & Keller, to appear). Acknowledgements: The following people have been involved in the previously unpublished results reported here: Pia Knöferle, Sandra Muckel, and Christoph Scheepers.

References:
G. Altmann & Y. Kamide (1999) Incremental interpretation at verbs: Restricting the domain of subsequent reference. Cognition, 73, 247-264.
M. Crocker and T. Brants. Wide Coverage Probabilistic Sentence Processing. Journal of Psycholinguistic Research; 29(6):647-669, 2000.
M. Crocker and F. Keller (to appear). Experience-based models of human sentence processing. Trends in Cognitive Sciences.
D. Jurafsky (1996). Probabilistic Model of Lexical and Syntactic Access and Disambiguation, Cognitive Science, 20:137-194.
M. Tanenhaus, M. Spivey-Knowlton, K. Eberhard & J. Sedivy (l995). Integration of visual and linguistic information in spoken language comprehension. Science, 268:632-634.
Y. Kamide, C. Scheepers, G. Altmann & M. Crocker. Integration of syntactic and semantic information in predictive processing: Anticipatory eye-movements in German. Paper presented at the CUNY Conference on Sentence Processing, New York, NY, March 2002.

Tim Dalgleish
MRC Cognition and Brain Science Unit, Cambridge (UK)
Emotions and inhibitory control

This talk will consider various ways in which aspects of emotion are subject to inhibitory control. A 3 by 2 taxonomy will be presented in which either automatic or controlled processes of inhibition are applied to emotional feelings, emotional cognitive content and emotional behavior(s). This will be illustrated with everyday and clinical examples. For example, psychogenic amnesia following a distressing experience might be viewed as the automatic inhibition of emotional cognitive content.

Manuel Garcia Carpintero
Department of Philosophy, Barcelona (Spain)
Qualia that It Is Right to Quine

Dennett (1988) provides a much discussed argument for the nonexistence of qualia, as conceived by philosophers like Block, Chalmers, Loar and Searle. My goal in this paper is to vindicate Dennett’s argument, construed in a certain way. The argument supports the claim that qualia are constitutively representational. Against Block and Chalmers, it rejects the detachment of phenomenal from information-processing consciousness; and against Loar and Searle, the argument defends that qualia are constitutively representational in an externalist understanding of this. Before providing the argument, I contrast firstly a minimal conception of qualia, relative to which their existence is not under dispute, with the sort of view to which I will object; secondly I set the stage by presenting the facts about (minimal) qualia on which a Dennett-like argument can be based.

References:
Block, Ned (1995): «On a Confusion about a Function of Consciousness», Behavioral and Brain Sciences 18, 227-247.
Chalmers, David (1996): The Conscious Mind, Oxford: Oxford University Press.
Dennett, Daniel (1988): «Quining Qualia,» in T. Marcel and E. Bisiach (eds.), Consciousness in Contemporary Science, Oxford: Oxford University Press, 42-77.
Loar, Brian (1997): «Phenomenal States», in N. Block, O. Flanagan & G. Guzeldere (eds.) The Nature of Consciousness, Cambridge, Mass.: MIT Press, 597-616.
Searle, John (1992), The Rediscovery of the Mind, Cambridge, Mass.: MIT Press.

Peter Gärdenfors
Lund University Cognitive Science (Sweden)

Concept learning in geometrical spaces

Traditional theories of concept formation, such as symbolic or associationist representations, have problems explaining the quick learning exhibited by humans. In contrast to such models, I advocate a third form of representing information that employs geometrical structures. I argue that this form is appropriate for modelling concept learning. By using the geometrical structures of what I call conceptual spaces, one obtains a general way of modelling similarity judgments. I define properties and concepts in term of conecptual spaces. A learning model that shows how properties and concepts can be learned in a simple but naturalistic way is then presented. I also discuss the advantages of the geometric approach over the symbolic and associationist traditions.

References:
P. Gärdenfors: Conceptual Spaces: The Geometry of Thought, MIT Press, Cambridge, MA, 2000.
P. Gärdenfors: "Concept learning: A geometrical model", Proceedings of the Aristotelian Society, vol. 101 (2001), pp. 163-183.

 SYMPOSIA
 

CONSCIOUSNESS OF ACTION AND SELF-RECOGNITION SYMPOSIUM
Convenor : Marc Jeannerod (ISC, France)
Speakers : Shaun Gallagher, Gunther Knoblich, Thomas Metzinger

The general framework of the session will be the following : What are the determinants of our consciousness of action ? And how does consciousness of action contribute to self-recognition ? Experimental as well as theoretical contributions will be presented. The levels of consciousness of action will be explored. At the lower level, automatic, non-conscious, mechanisms operate for visuomotor adaptation. This level only requires the integrity of basic mechanisms for the sensory control of action. At a higher level, a mechanism of monitoring “how to do” operate. Its existence is demonstrated by the shift in strategy which occurs when the automatic level becomes insufficient for achieving an action. At still a higher level, another conscious mechanism comes into play for assigning the action to its real origin. This function requires, not only a simulation of the whole action and its consequences on the external world, but also the monitoring of intention-related signals. This is a key mechanism for correctly attributing intentions to their authors. The attribution of intentional states to other people would thus be based on our ability to monitor our own mental states, so that, if an intention is not recognized as ours, it must pertain to someone else.

Shaun Gallagher
Department of Philosophy and Cognitive Science Canisius College, Buffalo (USA)

Self-agency and self-ownership

The distinction between the sense of ownership and the sense of agency is both useful for understanding some important aspects of self-consciousness, and clarifying for an analysis of symptoms of thought insertion and delusions of control in schizophrenia. Graham and Stephens (1994) make such a distinction in their analysis of these symptoms. I argue, however, that their account places the distinction at a level that is too "high" in the cognitive activity of self-consciousness - in effect, they locate it at a second-order level that involves the subject's reflective attribution, report, or introspection. Although John Campbell (1999) and Christopher Frith (1992) do not make a clear distinction between sense of ownership and sense of agency, their analyses of these symptoms appear to be targeted at that same level.
The experimental science suggests that such symptoms originate at a "lower" or first-order level of self-awareness. The difference in levels corresponds to the difference between a higher-order thought (HOT) theory of consciousness (e.g., Rosenthal) and a theory of consciousness based on the implicit structure of (pre-reflective) self-awareness. It is also reflected in the difference between a theory-of-mind approach to one's own intentional states and a view that self-awareness can be non-conceptual (e.g., Bermudez).
On the basis of neuroscientific evidence I argue that for understanding the origin of symptoms of thought insertion and delusions of control the proper distinction between sense of ownership and sense of agency is located at the level of implicit non-conceptual self-awareness. A disturbance at that level is what motivates problems at the higher level of cognitive reflection.
 
 

Günther Knoblich
Max Planck Institute, Munchen (Germany)

Self-recognition of earlier actions

One aspect of self-recognition that has received little attention in prior research is the question of whether and how people recognize their own earlier actions. We have developed several experimental approaches to address this question. The results show (1) that people are able to recognize their earlier actions even when there is no cue to the identity of the person carrying out the action, (2) that they are better able to predict future results of their earlier actions. These results are best explained by the assumption that the action system contributes to action perception. This contribution might take the form of a concurrent internal simulation of observed actions that is mediated by a representational level coding distal events.
 

Thomas Metzinger
Department of Philosophy of the Johannes Gutenberg Universität Mainz (Germany)

Self-recognition and the gradient of phenomenal transparency

My contribution has three short parts. First, I will take a closer look at a concept, that has been too-much neglected, namely the notion of „transparency“: What precisely is „phenomenal transparency“? In the second part I will refine the standard notion as it is today used by philosophers like, e.g. Shoemaker, van Gulick, Harman, Mandik or Tye by imposing additional phenomenological and teleofunctionalist constraints.. In particular, I want to draw attention to a simple  phenomenological fact, the fact that phenomenally opaque states do exist. I then proceed to ask: What does it mean that there is a gradient of transparency in the human self-model? I will close by offering an application of this new conceptual differentiation to self-recognition. I hope to show that the gradient of transparency does affect the process of self-recognition, functionally as well as phenomenologically.
 

MOTOR INTENTION AND THE CONTROL OF ACTION SYMPOSIUM
Convenors: Y. Rossetti and E. Pacherie
Speakers: H. Bekkering, E. Pacherie, L. Pisella, I. Toni

This symposium aims at reviewing recent progresses made by philosophers and neuroscientists on the level of implication of intention on action control. Four contributions are proposed. A philosopher (E. Pacherie) will contrast two aspects of the online control of action : present-directed intentions and motor intentions can be distinguished on the basis of their time scale and representational level. A brain imaging contribution (I. Toni) will describe recent sophistications of delayed tasks used to distinguish the networks involved in movement preparation, action regulation and motor intention. A psychological approach (H. Bekkering) will show how imitation can be viewed as a tool to analyse the mechanisms and the goals of intentional action control. A neuropsychological account (L. Pisella) will argue that intention and automatism can be viewed as the two poles of a gradient where mediate responses constitute progressive steps towards intention.

Elisabeth Pacherie
Institut Jean-Nicod, Paris (France)
On line control of action

My purpose is to contrast two levels of online control of action: online rational control and online motor control, associated respectively with present-directed intentions (P-intentions) and motor intentions (M-intentions). I characterize each level in terms of its dynamics, of the temporal and rational constraints it is subject to, and of the type of representations it exploits. I also discuss the way the two levels of control are related and the limits to their compatibility.
A P-intention is an intention to act now whose task is to anchor an action plan in the situation of action. The temporal anchoring, the decision to start acting now, is but the beginning of this process. Having established a perceptual information-link to the situation of action, the agent must insure that the action plan gets implemented in that situation. This means that he must effect a transformation of the purely descriptive contents of the action plan into indexical contents anchored on the situation and control the execution of the plan. The rational control of an action by a P-intention takes two forms. Tracking control enables an agent to keep track of her way of accomplishing an action and to adjust what she does so as to maximize her chances of success. Collateral control is control by the agent for side effects of her way of accomplishing an action. Both types of control are rational insofar as what the agent does in both cases is connect her indexical conception of her ongoing action to general principles of means-end reasoning, to her desires, values, general policies and rules of conduct. The agent exercises rational control over her action insofar as (1) she is in a position to judge whether or not her way of accomplishing her action is likely to lead to success and to adjust it so as to maximize her chances of success (tracking control) and (2) she is also in a position to judge whether or not it brings about undesirable side-effects and to correct it accordingly (collateral control). P-intentions are thus responsible for high-level forms of guidance and monitoring applying to aspects of the situation of action and of the activity of the agent that are both consciously perceived and conceptualized. Their time scale is therefore the time scale of conscious perception and rational thought.
This level of rational control is to be contrasted with much more fine-grained forms of control of an ongoing action, that are responsible for the precision and speed of execution and whose time scale is that of a neurological micro-present that only partially overlaps the conscious present. Whereas rational online control is a function of P-intentions, fine-grained motor control is a function of M-intentions. The latter involve a transformation of the perceptual indexical content of P-intentions into motor representations. We are therefore moving to a second level of situational anchoring that is not simply perceptual but sensory-motor. Motor control makes use of internal models of action that capture the relationships between intended sensory consequences and motor commands yielding those consequences (inverse models) and that predict the sensory consequences of motor commands (forward models). Motor control is not subject to the strong rationality constraints characteristic of higher-level control. Motor control systems are at least modular to some degree: they have limited access to information from other cognitive systems or subsystems and our conscious access to their operation and the informational content they exploit is also limited.
Finally, I argue that although conscious intentional action requires a transformation of conceptual and perceptual representations into sensory-motor representations, differences in time-scales as well as differences in the representational formats used and the types of information encoded allow for discrepancies between rational and motor control.

Ivan Toni
F.C. Donders Centre for Cognitive Neuroimaging (The Netherlands)

Movement preparation and motor intention

Flexible goal-oriented behaviour relies on spatial coordinate transformations and on motor control mechanisms, but also on the capability to take advantage of contextual information for steering the sensorimotor machinery. While accurate performance of a sensorimotor task requires parietal and frontal regions, their differential contribution and their functional relationship with other associative regions remains obscure.
We have used event-related fMRI to measure human cerebral activity associated with motor cognitive processes in the context of delayed performance of an associative visuomotor task. Movement instruction (specified by visual cues) and motor performance (specified by an auditory cue) were separated by a variable delay period. By manipulating the predictive value of the instruction cue we distinguished delay-related activity influenced by response probabilities (motor preparation and motor inhibition) from delay-related activity un-affected by the likelihood of providing a motor response (motor intention). Furthermore, such manipulation allowed us to challenge not only the interplay between spatial frames of reference and kinematic control processes, but also the extraction of contextual information from the task contingencies.
We found delay-related activity distributed across a cerebral network involving not only frontal circuitry, but also extrastriate and medio-temporal regions. Areas showing motor intentions and preparatory responses were spatially intermingled. Posterior parietal cortex deviated from this pattern, showing delay-related activity irrespective of movement probability, but no specific preparatory responses.
Our data suggest that posterior parietal cortex and dorsal precentral cortex might play different strategic roles in solving these visuomotor problems. While posterior parietal cortex covers a range of potential responses defined by the task setting, dorsal precentral cortex focus on a probable movement. We speculate that a distributed temporo-prefrontal system gathers information on such potential and probable responses on the basis of the trial-by-trial contingencies. Finally, it might be noticed that prefrontal and temporal regions, although supposedly involved in regulating behavioural responses during task performance, operated independently from overt verbal reports, a sign of awareness of regulation of actions.

L. Pisella and Y. Rossetti
Espace et Action, Inserm, Lyon (France)

Mediate responses as cues to intention: a neuropsychological approach

 In the field of neuroscience, Clinical neurology and Neuropsychology, Experimental psychology, and Motor psychophysics have shaded different lights on intentional processes. The clinical approach traditionally focuses on the alteration of intentional control at the level of executive functions, as they typically appear following frontal lesions. Experimental psychology has been focusing on the analysis of reaction time and on inhibitory mechanisms. Motor psychophysics has explored the way motor control can be dissociated from awareness of external events. Combinations of these approaches have proved to be most productive. In particular, the exploration of neurological patients by psychological or psychophysical means have shown that intentional control can be dissociated from automatic control of action. It appears that most experimental conditions used in animal as well as in human experiments explore stimulus-triggered reactions rather than self-initiated responses. Therefore relevant indirect responses are reviewed here and proposed as tools for exploring intentional aspects of behaviour. Further experiments have suggested that the dissociation between intentional vs. automatic motor control can be viewed as a functional interpretation of the ventral and dorsal anatomical pathways. It is here rather argued that it correspond to a parietal/frontal distinction. The most appropriate view of the organisation of the visual-to-motor network seems to be a functional and a temporal gradient of motor responses. One pole of this gradient is easily identifiable, and seems to deal exclusively with the automatic on-line control of direct goal-directed actions. The other pole could be intention, supported by a much richer anatomical network (not to react can be considered as the most obvious manifestation of intention in repetitive experimental conditions). Between these two poles, a range of indirect processing – where the most direct response to a stimulus has to be inhibited or overcome – can be described, that provide a cue to intentional control of a simple action. Going beyond the classical dichotomy between a dorsal and a ventral route to action, these tasks suggested that more than two functions can be distinguished within the motor repertoire. What and where is intention within this network remain open questions to philosophers and neuroscientists.
 
 

Harold Bekkering
Department of experimental and work psychology (The Netherlands)
Imitation: A tool to study the underlying mechanisms of intentional action control

Currently, imitation, or performing an act after perceiving it, is in the focus of attention of researchers from many different disciplines. First, an overview of common theories about imitation will be given. Then, results of recent experiments that used an imitative design (you do as I do) will be presented. These experiments were designed to study the organization of action control in children, healthy adults and patients with Ideomotor Apraxia. In addition, neurophysiological information about the processes underlying mechanisms in imitation by using fMRI will be offered. Together, the results suggest that the key concept to understand imitative behavior is formed by action goals. We propose that two sensorimotor mechanisms are involved in the execution and the perception of action goals in imitation, respectively. First, the ideomotor principle--selected physical goals like an object to grasp elicit the motor program with which they are most strongly associated, elucidates how both the agent and the imitator translates action goals into the execution of movements. Second, the mirror neuron system--matching the observation of goal-directed movements with the individual motor actions, gives details of how seeing goal-directed movements can give rise to the recognition of action goals. Thus, the recognition of action goals from the observation of movements and the execution of movements from having action goals can be conceptually clarified by proposing two inverted functional mechanisms. The mirror neuron system transforms movements perceived from others into possible actor’s own action goals, while the ideomotor principle transforms the intended action goals of an actor into movement execution. To conclude, the functional model of imitation as described here, stresses the importance of the personal action repertoire in both the perception and the execution of goal-directed imitative actions.
 
 
 

PRAGMATICS SYMPOSIUM
Convenor : Richard Breheny (University of Cambridge)
Speakers : Robyn Carston, Bart Geurts, Charles Travis
 
 

Bart Geurts
Department of philosophy, University of Nijmegen (The Netherlands)

Interpretation without meanings
 

One of the principal schools of semantics (as well as certain parts of pragmatics) is the dynamic approach to interpretation, which was launched by Hans Kamp and Irene Heim at the beginning of the eighties. The dynamic framework has spawned analyses of a broad range of phenomena, including anaphora, presupposition, indexicality, quantification, plurality, tense and aspect, modality, and propositional attitudes.The main objective of this talk is to elucidate some general implications of the dynamic approach with respect to the concept of meaning and the architecture of a theory of interpretation. Hence, for the purpose of this talk I will assume that the dynamic approach is essentially correct. I will work with one particular species of dynamic theory, viz. discourse representation theory (DRT).
As its name already indicates, DRT is concerned with the interpretation of discourses, not just sentences. As a discourse unfolds, interlocutors incrementally construct a model of what has been said. Such models are called "discourse representation structures" (DRSs). DRSs may be viewed as representations of commitment slates in the sense of Hamblin. A speaker who utters a sentence S makes a commitment of sorts. He need not actually believe that S is true, but he is committed to act, at least for the time being, as if he believed S to be true. We may imagine that each speaker has a slate on which his conversational commitments are recorded, and the interlocutors maintain representations of each other's commitment slates. In DRT such representations are taken to be DRSs. In my version of DRT, DRSs are built in two stages. In the first stage, a DRS is constructed which is determined compositionally ‹ i.e., it is a function of the lexical contents and the syntactical constructions that enter into the grammatical analysis of the sentence. By the time this level of representation is reached various kinds of ambiguity have been resolved, but a sentence DRS still leaves open the interpretation of pronouns, presuppositions, tense, and so on. Such matters are dealt with in the second stage of interpretation, when the sentence DRS has been merged with the DRS
that represents the preceding discourse. This view on interpretation has a number of general consequences, of which I want to discuss the following three:
Propositions:
In this account the traditional notion of proposition does not figure in any way. In a sense, DRT doesn¹t really provide an answer to the question what a sentence, as uttered on a given occcasion, means. On the one hand, sentence DRSs are so schematic that they barely qualify as semantic representations at all. On the other hand, once a DRS has been "fleshed out" (i.e. anaphoric references have been resolved, presuppositions have been taken care of,
etc.), it has meshed with the representatiof the previous discourse, and it is impossible to separate the utterance¹s own contribution from contextual information. This means that the traditional notion of proposition has to go.
Compositionality:
The DRT account is blatantly non-compositional: to the extent that we can speak of meanings at all, the meaning of a complex expression is not determined by the meanings of its parts and the way they were put together. Thus, DRT proves that a non-trivial theory of interpretation need not comply with the compositionality principle.
Early vs. late pragmatics:
Although, strictly speaking, DRT does not in fact commit itself one way or the other, it is natural to assume that at least some pragmatic inferences belong to a third stage in the process of interpretation. For example, various implicatures would be computed at this stage. Thus we arrive at a architecture that, broadly speaking, is in line with the general picture Grice had in mind, though it deviates from the letter of Grice's writings in various respects. In particular, it is doubtful that our distinction between early and late pragmatics is a very deep one, e.g. in view of the fact that bridging inferences (which do not seem to be different in kind from some
classic implicatures) presumably are part of early pragmatics.
 

Robyn Carston
Department of Phonetics and Linguistics University College, London (UK)

Linguistic Underdeterminacy and Pragmatic Enrichment

While the linguistic underdeterminacy (or underspecification) of utterance meaning is widely recognised nowadays within semantics and pragmatics, there are a number of  different ways in which it can be understood depending on (1) what it is that is taken to be underdetermined, and (2) what it is that does the underdetermining (that is, what ‘linguistic meaning’ is).
Three possibilities for (1) are:
(a) full utterance meaning (explicit content and implicatures),
(b) just what is explicitly communicated,
(c) the proposition expressed, which may not be communicated (speaker-intended).
Three possibilities for (2) are:
(a) context-free linguistic expression-type meaning,
(b) context-relative, but speaker intention-free, linguistic expression meaning,
(c) context-relative, and possibly speaker intention-relative, linguistic meaning.
Working within the relevance-theoretic approach to utterance understanding, developed by Dan Sperber and Deirdre Wilson (1986/95), the version of the underdeterminacy thesis that interests me concerns what is explicitly communicated (or ‘what is said’, on some construals of that term), i.e. (1b).  The approach comes with certain architectural commitments: there is an autonomous (modular) language decoding system and a distinct pragmatic comprehension system responsible for the full range of pragmatic tasks, including disambiguation, saturation of indexicals, and implicature derivation.  It follows, then, that the linguistic meaning at issue here is that in (2a), i.e. ‘context-free expression-type meaning’ and that pragmatic inference plays a major role in bridging the gap between linguistic meaning and explicit utterance content.
I will argue that there is no portion of utterance meaning that meets the description of linguistic meaning in (2b), notwithstanding arguments about pure indexicals and objective contextual parameters (e.g. speaker, addressee, time and place of utterance).   As regards (2c), on the other hand, there certainly is an isolable part of utterance meaning that can be so described; it precisely is the explicature (or ‘what is said’, on some construals) of the utterance, so that this definition of linguistic meaning renders false the underdeterminacy thesis of interest: linguistic meaning (2c) does determine what is explicitly communicated (1b).  However, on the favoured relevance-based approach to comprehension, this definition of linguistic meaning cuts across the cognitive processes involved, since explicit content is a composite of linguistic decoding and certain processes of the pragmatics module, governed by the same non-linguistic principles which are responsible for the derivation of implicatures.  Furthermore, if, as will be argued, some of the pragmatic processes contributing to explicit content are ‘free’ (i.e. not linguistically mandated), there seem to be no grounds for this conception of ‘linguistic’ semantics, leaving us just with (2a), expression-type meaning.
This surviving notion of linguistic meaning does not meet the usual requirement of propositionality and so, as many semanticists would claim, cannot bear the crucial semantic relations of reference and truth.  This is quite unsurprising on the (necessarily psychologically solipsistic) account of utterance understanding, in which the function of linguistic forms is to guide addressees in inferring the communicator’s intended meaning.  It is propositional mental representations (thoughts), including those that constitute the interpretation of an utterance, that stand in the ‘genuine’ (or primary) semantic relations with such nonrepresentational entities as objects, properties and events.
 
 

GAME THEORY AND THE SEMANTIC/PRAGMATIC INTERFACE SYMPOSIUM
Convenor : Robert van Rooy (University of Amsterdam)
Speakers : Arthur Merin, Nicholas Asher, Prashant Parikh
 

There exists a gap between the semantic representations of sentences and the thoughts actually communicated by utterances. How should this gap be filled? The obvious answer (Grice, 1957) seems to be that the hearer should recognize what the speaker thinks that the listener understands. Because this depends in turn, in a circular way, what the listener thinks that the speaker has in mind, PRASHANT PARIKH will argue that a game theoretical framework seems natural to account for such situations. Intuitively, what goes on here is a game between a speaker and a hearer where the former chooses a form to express the intended meaning, and the latter chooses a meaning. But how can a hearer recognize the speaker's intentions? Gricean pragmatics (1975) suggests that she can do so by assuming that the speaker is cooperative and thus obeys his conversational maxims. Sperber & Wilson (1986) have argued that these 4 maxims can be reduced to the single principle of optimal relevance. But this gives rise to at least two questions: (i) can we simply assume that rational speakers are always cooperative, and even if so (ii) how could these maxims evolve? ARTHUR MERIN will argue that communication is guided by the principle of optimal relevance, but that this notion of relevance should be thought of from an argumentative point of view, rather than a purely cooperative one. Merin will use game theory to formulate his principle of relevance. NICHOLAS ASHER will address the question of how maxims of conversation, or default rules of interpretation, could arrive. He will argue that some of these maxims can be explained as stable solutions of evolutionary games.
 
 

Nicholas Asher
University of Texas, Austin (USA)

Games underlying defaults for discourse interpretation

In this talk I will present a picture of discourse interpretation on which default rules of interpretation play an important role. Some of these defaults have to do with speaker attitudes; more specifically, they concern between what speakers intend, desire or believe and what they say. The question is how do such defaults arise? I explore the connection between game theory and these defaults.
 

Prashant Parikh
University of Pennsylvania (USA)
Language and Information Flow

I will use game theory to show how the distinction between natural and nonnatural meaning (or information flow) that Grice introduced in the late forties can be understood and defined. It will be shown how the distinction gives rise to a range of concepts of information flow and how game theory provides a neat way to capture this range. In so doing, it will be seen that it is necessary to go beyond the traditional confines of game theory and consider more general structures that I call strategic interactions. I will also touch upon how game theory can be used to model ambiguities in language as well as implicatures.
 
 

Arthur Merin
Department of Philosophy, University of Konstanz (Germany)

Language games on objects of desire: Prospect semantics

Theories of rational agency and imputable mind are centrally constituted by constraints on agents' beliefs and desires. The objects of belief are propositions, i.e. elements of Boolean or, on some accounts, slightly more general Heyting algebras. The objects of desire are, once again, propositions to any philosopher who treats desire as a propositional attitude. Yet in the standard decision and game theories of economists and psychologists, the objects of desire are something quite different, namely prospects. These may be risky, and thereby functions of beliefs; or riskless, and thus formally independent of beliefs. But either way, and by strictly algebraic criteria, in these theories, prospects are not propositions; nor propositional property abstracts. Indeed, the elegant assumption of a single proposition algebra of objects of belief and of desire is well known to invite, in some contexts, deviations from standard game-theoretic rationality by leading, in the absence of additional constraints, to the recommendation, as expected value optimizing, of strictly dominated strategies. Having beliefs does not a priori presuppose having desires. Traditional recursive theories of meaning thus make objects of belief, i.e. propositions, their basic building blocks without assuming anything about desires. Having desires, in return, does not a priori presuppose having beliefs. Theories of riskless choice accordingly proceed without making any assumptions about beliefs. But then we should ask in all fairness: Can objects of desire be the building blocks for a recursive theory of meaning and of associated concepts?
We show they can, by giving semantics in riskless prospects for sentential languages in the imperative mood. Prospects are elements of ordered group algebras, their product spaces, and linear algebras extending them. Their formal ontology underlies a paradigmatic two-person game of perfect though incomplete information. The ontological and game structures are shown jointly to account for robust empirical aspects of meaning which are not, so far, otherwise accounted for.
 
 

HYPNOSIS SYMPOSIUM
Convenor: Zoltan Dienes
Speakers: Zoltan Dienes, Paloma Enrique and Sakari Kallio

This symposium will address the nature of hypnotic responding. The first talk, by Zoltan Dienes, looks at the relation between representational explicitness, on the one hand, and both consciousness and control, on the other. By arguing that consciousness requires a greater degree of explicitness than executive control, the possiblity is open for the counter-intuitive case of unconscious executive control. And Dienes suggests this is exactly what hypnosis is.
One issue that has aroused strong feelings over the decades is whether a special state need occur to provide the background context for such hypnotic responding. Kallio and Revonsuo and Enrique and Poppel consider the possiblity of a change in neurophysiological mechansims that do not have mental content themselves, but modulate the functioning of states that do have content. Could such accounts provide the basis for explanatory theories postulating altered states of consciousness (ASC)? Or does hypnosis simply use mechanisms that are part and parcel of everyday psychology?
 
 

Zoltan Dienes
Laboratory of Experimental Psychology School of Biological Sciences, University of Sussex (UK)
Explicit knowledge, higher order thoughts, and hypnosis.

In this talk, I will describe a hierarchy of ways by which knowledge can be explicit (Dienes & Perner, 1999). The successive levels of the hierarchy are called: property explicit, predication explicit, fact explicit, and finally attitude explicit. In the proposed hierarchy, the most explicit sort of knowledge (attitude explicit) is conscious knowledge by the Higher Order Thought theory of Rosenthal (2002). The different levels also relate to different types of control, as argued by Perner (1998, 2002): Vehicle control (contention scheduling) can use fact implicit representations, but content control (supervisory attentional system control) must use fact explicit representations. I will argue that hypnosis highlights a situation in which content control is used with attitude implicit representations, and it is this dissociation between fact and attitude explicitness that gives hypnosis its bizarre feel (executive function responding in the absence of conscious intentions). The arguments are consistent with the theories of Spanos and Hilgard from the 80s, but inconsistent with more recent theorizing in hypnosis by Kirsch, Lynn, Bowers and Woody.

References
Dienes, Z., & Perner, J. (1999) A theory of implicit and explicit knowledge. Behavioural and Brain Sciences, 22, 735-755. Available at: http://biols.susx.ac.uk/Home/Zoltan_Dienes/publications.html
Perner, J. (2002). Dual control and the causal theory of action. In N.Eilan & J.Roessler (Eds.). Agency and self-awareness. Oxford: Oxford University Press. Available at:  http://www.sbg.ac.at/psy/people/perner/literatur_e.htm#buecher
Rosenthal, D. M.(2002). "Consciousness and higher order thought", Macmillan Encyclopedia of Cognitive Science. Available at: http://web.gc.cuny.edu/cogsci/dr.htm#publications
 

Sakari Kallio and Antti Revonsuo
Department of psychology, University of Turku (Finland)
Towards an empirically testable account of hypnosis - Altered state of consciousness or social interaction?

There is currently wide agreement that, in addition to the changes in external behavior, suggestions presented in a hypnotic context may give rise to changes in subjective experience. Yet, there is no general agreement about the theoretical framework within which they should be explained. The disagreement focuses on whether reference to a specific internal state of the individual is necessary in order to explain these changes. We place the explanatory task in the context of mechanistic explanation which reveals that the disagreement between the "state" and "non-state" view is about the level of description at which the phenomenon "hypnosis" should be conceptualized. After this analysis, we propose a novel approach using the model of mechanistic explanation which has proved to be very useful in biological sciences. This kind of approach clarifies the explanatory task of hypnosis research and helps to formulate empirically testable hypotheses about the nature of hypnosis. We will outline the basic elements of such an approach, and hope that our proposition and integrate hypnosis research with the currently increasing multidisciplinary research on other phenomena of consciousness.
 
 

Paloma Enríque and Ernst Pöppel
UNED University, Madrid (Spain)

Towards a comprehensive neural view on psychological phenomena : separating disctinct brain functional domains

Scientific work is currently characterized by a deficit in new theoretical developments in which recent empirical contributions could be 'explicitly' framed. Consequently, results of research are being frequently located into an 'implicit theoretical frame' grounded on non re-elaborated previous ideas.
Such an state of things is expressed in cognitive neuroscience by a number of implicit assumptions, among them are: (1) Between brain and psychological functions there should be a close correspondence; (2) The main functional neural code is topographical; and thus, (3) Every psychological function has to be specifically brain located.
Against these usual implicitly assumed preconceptions, it is of great theoretical interest to distinguish between two broad kinds of functional brain domains:
- 'What functions'.- Those brain operations which provide the specific psychological contents leading to our subjective conscious experience, such as perceptive, mnemonic, and linguistic ones.
- 'How functions'.- Those very different 'psychologically content free' brain operations which support every specific what activity, as activation, temporal organization, attention, and integration, for instance.
Differential properties of these two distinct kinds of brain functional domains will be analysed, as well as the relevance of such a distinction for reaching a comprehensive conciliated view on several fields usually treated from very distinct perspectives, as neuropsychology, psychopathology, and consciuousness studies.
 
 
 

PARALLEL SESSIONS

Erkki Ahlström
Institut des sciences cognitives, Lyon (France)
Semantics-pragmatics interface: The referential-attributive distinction revisited. Some linguistic evidence concerning its semantic status

In this paper I try to show that in certain situational and textual uses, where definite description (the F)  and complex demonstratives (that/this F) follow certain patterns of preference, the semantics of definite descriptions developed by Russell and defended by a number of his followers, e.g., Stephen Neale, provides a viable explanation to the distribution of determiners. Empirical observations point in the direction of contextual features predicting the preference patterns between complex demonstratives and definite descriptions. Like other indexical  expressions, complex demonstratives function as designators in the strong sense, which means that from a semantic point of view they introduce the referent to the truth conditions of the sentence in  which they appear. On the other hand, definite descriptions are quantificational expressions. They are the most relevant way to identify the referent in certain. Circumstances. Moreover, as the quantificational features of definite descriptions seem to function in a normal way also in referential cases, this in turn can be considered as evidence supporting unambiguous Russellian semantics and as falsification of the semantic ambiguity thesis of definite descriptions.
 
 

Nicholas Allott and Paula Rubio Fernández
Research centre for English and applied linguistics, Cambridge University (UK)
This paper fills a much-needed gap

In the ‘Moses illusion’, a case of shallow discourse processing, people tend to answer the question How many animals of each kind did Moses put in the ark? without noticing the error (Erickson & Mattson, 1981). The results of a number of studies suggest that the ‘global goodness of fit’ of a term in a given mental scenario may influence the level of semantic analysis of this term. Barton and Sanford (1993) investigated this hypothesis in detail and interpreted their results according to the ‘scenario-mapping and focus account of text comprehension’ (Sanford & Garrod, 1981). Although offering a consistent interpretation of the Moses illusion phenomenon, their view of text comprehension seems too limited to account for pragmatic illusions where the anomalous term is not expected in the given scenario but rather relevant in the context. In this paper, we argue that a relevance-theoretic approach provides a good account of pragmatic illusions, and offers an explanation for the different levels of processing and accessibility of semantic information in text comprehension.
 
 

Bacon E. , Paire-Ficout L. , & Izaute M.
Inserm U 405, Clinique Psychiatrique, Strasbourg (France)
Dissociation between subjective experience and the cognitive process under the effect of lorazepam : the case of the TOT state

The tip-of-the-tongue state (TOTs) refers to the subjective feeling that a particular target word is on the point to be retrieved. It corresponds to a transitory state of inaccessibility of a known information. Diary studies and laboratory tasks show that 50 to 70% of the TOTs are accompanied by intrusive blocking words. These blockers are recognized as incorrect, but subjects cannot retrieve the correct target. When rememberers experience TOTs, they are more likely to retrieve eventually the correct answer. A TOT reveals a conflict between the metacognitive judgement, i.e. the certaincy that the information is known, and the cognitive level, i.e. the momentaneous inability to retrieve the target from long term memory. TOTs are considered as slowing down of a memory process, and they may be viewed as momentaneous and reversible « microamnesia » that naturally occur in healthy subjects.
On the other hand, very few drugs have been shown to impair semantic memory.  With a task consisting of questions on general knowledge that assessed semantic memory, we observed that Lorazepam (LZ), a benzodiazepine, impaired the performance of recall, as subjects provided as many responses as the subjects under placebo, but with more incorrect answers (commission errors). In addition, LZ-treated subjects overestimated the correctness of their wrong answers (Bacon et al, 1998, Massin-Krauss et al, 2002). In this study, we subsequently explored experimentally the TOT hypothesis as a possible explanation for this temporary semantic impairment induced by lorazepam. We investigated the effects of the benzodiazepine lorazepam on the TOT states and on the subsequent memory performance. Thirteen lorazepam (0.038mg/kg) and 13 placebo subjects were examined using a semantic memory task. The results showed that more TOTs were recorded after incorrect recalls for the lorazepam group than for the placebo subjects, whereas a similar amount of TOTs was recorded after omissions (no answer) in both groups. The overall resolution of the TOTs (to recognize the correct target after having experienced TOTs) was preserved. It seems that the higher amount of incorrect recalls provided by subjects under lorazepam could partially result from the fact that they are more frequently in a state of transitory inaccessibility of a known item than placebo subjects, which is experimentally identifiable as a commission TOT :  the subjects know the answer, but they provide a blocker that obstructs the access to the correct answer. However the subjects do not spontaneously feel the phenomenology of the TOT, and they do not recognize this response as an impostor. Thus, lorazepam seems to impair the relationship between the subjective experience and the cognitive process involved in the TOT experience. This peculiar way of resolving the TOT conflict is discussed in the light of the anxiolytic effect of the drug.
 

M. J. Bergman - A. Revonsuo - S. Kallio
Institution of Humaniora (Sweden)

A research model of hypnosis from the viewpoint of mechanistic explanation

 The theoretical debate about the nature of hypnosis has settled around two different camps: special process or altered state view (SV) and a social psychological or nonstate view (NSV). SV-theorists argue that hypnosis may lead to an altered state of consciousness and various brain imaging methods are used to reveal the neural correlates associated with different hypnotic phenomena. NSV  theorists argue that hypnotic phenomena are due to social factors involved in the situation. Changes in behavior and even subjective experiences are explained by social factors. The theoretical debates have continued for at least 30 years in different forms, without any resolution in sight.  In order to analyse the assumptions involved in this debate, we place the explanatory task in the context of mechanistic explanation, an explanatory framework often used in the life sciences and recently much explored in the philosophy of science. Mechanistic explanation implies that complex biological phenomenon consists of different hierarchically organised levels and a complete explanation of a phenomenon should include all levels that are relevant for it. Mechanistic explanations approach the phenomenon from different directions: etiological explanation (backward looking), constitutive explanation (downward looking) and contextual explanation (upward looking). These explanations do not consist of propositional theories, laws or deductive arguments, but are instead typically presented as graphical representations and visualizable models that include several levels of description. We will present an outline of a research model of hypnosis from the viewpoint of mechanistic explanation. In this outline hypnosis is regarded as a complex biological phenomenon consisting of different levels of organisation. This kind of approach clarifies the explanatory task of hypnosis research and helps to formulate empirically testable hypotheses about the nature of hypnosis.
 
 

Bongioanni Paolo, Buoiano Giancarlo, Magoni Marzia
University of Trieste (Italy)
Linguistic impairments in amyotrophic lateral sclerosis/motor neuron disease
 

Motor Neuron Disease/Amyotrophic Lateral Sclerosis (MND/ALS) is characterized by progressive degeneration of the motor neurons within the central nervous system. Until recently, MND/ALS was thought to spare cognitive functions, but now it is acknowledged that about 3-5% of patients affected by MND/ALS have cognitive impairments.
In our work, we review 6 papers on language derangements in MND/ALS patients. Up to date, it is not possible to establish whether this subgroup is affected by MND/ALS and Fronto-Temporal Dementia (FTD), or whether the aphasia can be, in some cases, a consequence of MND/ALS itself.
Lexical category-specific impairments affect predominantly verbs, whereas nouns appear to be partially preserved. The deficit of verbs is remarkable, because it might shed light on the link between actions, verbal and cognitive imagery, and ideomotor praxia.
Language impairments detected in MND/ALS point largely to frontal and fronto-striatal derangements. Functional neuroimaging studies show reduced frontal activation. Alterations in Broca’s area, DorsoLateral PreFrontal Cortex (DLPFC) and fronto-striatal circuit have been reported in MND/ALS: several functional studies link Broca’s area to syntactic processing, and DLPFC to verbs production as well as language short term memory.
A basic issue is the link between action and verbs and, as a result, between action syntax and speech syntax. This connection would be involved in the origin of language: it would be evolved passing from action syntax to speech syntax. According to our point of view, verbs exert a pivotal role in this system. In short, when actions are missing, verbs together with the ideomotor praxia (and their neurocorrelates) would deteriorate.
A longitudinal study on the effect of massive motor derangement on the verb degeneration is presently carried out by our research group.

Keywords: MND/ALS, action, fronto-striatal circuit, syntax, verbs
 
 
 

Ingar Brinck
Department of Philosophy, Lund University (Sweden)
The pragmatics of imperative and declarative pointing.

The distinction between imperative and declarative pointing is based on behavioral and cognitive criteria. The notion of imperative pointing is fairly straightforward, while declarative pointing is characterised in different ways. An attempt is made to give a description of declarative pointing that is neutral in comparison to its more elaborated characterisations in terms of cognitive capacities. To this end, Bates et al.’s (1975) original account of the pragmatics of pointing is examined. The basic claim is accepted: Pointing is an illocution which requires the intentional use of a conventional signal to carry out some socially recognised function.
It is claimed that declarative pointing does not, pace Bates, have the same structure as imperative pointing. The function of declarative pointing is simply to indicate the presence of a particular object. Whether the kind of joint attention that is necessary for the success of an act of declarative pointing requires a theory of mind, or any other cognitive capacity that is under dispute, is a moot question that should be settled empirically. Joint attention does not in principle involve meta-representation, nor presuppose a theory of mind.
An illocutionary act may have different perlocutionary effects on, or consequences for, the actions and thoughts of the addressee depending on the context in which it occurs. It is claimed that any other interpretation of the function of declarative pointing than the one given here, either pertain to the enabling and preparatory conditions or relate to the perlocutionary effects of declarative pointing. Various kinds of manipulation of attention belong to the preparatory and enabling conditions. Obtaining attention, sharing experiences and emotions, or monitoring behavioral reactions towards the object are results of the perlocutionary effects of declarative pointing.
 
 

Justin Broackes
Brown University, Providence (USA)
Colour Blindness, Colour Language and the Dimensions of Visual Appearance

Dichromats apparently possess two types of cone in the retina instead of the normal three, and, on a standard view, their colour vision can therefore have only two dimensions rather than the normal three. The space representing hue, saturation and brightness collapses to a single plane: the dichromat supposedly sees no more than two hues, with variations in their saturation and lightness -- e.g. (as in Judd 1948) a yellow of ~575 nm and a blue of ~470 nm. What then is going on when dichromats use terms like 'red', 'green', 'orange' and 'purple', if they can really only see one yellow hue and one blue? And what is going on when (as in Jameson & Hurvich
1978) some dichromats even apply such terms correctly to the coloured caps in the Farnsworth D-20 test?
There are many factors to consider (lightness cues, rod inputs, seeing a surface from various angles as it reflects light of various spectral profiles), as well as studies of cases of unilateral colour vision deficits. I shall say something about the projection of 2-dimensional colour-data into 3-dimensional space, and about aspect-shift in colour-perception -- a special case of the 'interpretation' of ambiguous stimuli. A survey of a variety of philosophical views of the nature of colours (and in particular, surface colours) suggests that we may even be able to allow the dichromat to have a proper conception of the colours whose spectral equivalents he cannot in the normal way see. All of which suggests some new directions for experimental investigation, which, as a philosopher, I am sorry not to be qualified to perform.
 
 

Stephen Butterfill
St Catherine’s College, Oxford (UK)
Understanding belief

What is involved in acquiring and acting on beliefs? Standard accounts treat believers like vending machines. Coins dropped into a slot cause the machine to go into different states, which in turn cause it to dispense various products. Similarly, it’s standardly assumed that events cause changes in belief, which in turn cause people to act in various ways. This talk aims to show that standard accounts of belief are inadequate. Two objections are presented. First, standard accounts ignore the thinker’s involvement in acquiring and acting on beliefs. We know that thinkers are sometimes involved in acquiring beliefs because they can find it hard to accept manifest but unpalatable truths, and because they may irrationally refuse to accept an obvious truth. Second, standard accounts treat which contents someone’s beliefs have as a matter for a theorist to decide by observation, as if it were no concern of the thinker herself. But an adequate account of belief cannot ignore a thinker’s understanding of her own beliefs, because her seeing why she should acquire a belief sometimes explains why she does acquire it. In short, standard accounts treat ordinary people as uncritical repositories for belief, rather than reflective thinkers who are insightfully involved in acquiring and acting on beliefs.
 
 

Philippe Chuard
The research school of social sciences, Australian National University (Australy)
From Concepts to Appearances

The idea that what conceptual capacities a given perceiver possesses may either determine, or at least causally influence, the way her experiences represent her environment seems widespread. In philosophy, this intuition goes hand-in-hand with Conceptualism about perceptual content, according to which the representational content of perceptual experiences is wholly conceptual. I argue that no good argument for Conceptualism can be mounted on the basis of the various phenomena that seemingly support this intuition.
 
 
 

Elena Daprati, Daniele Nico, Arnaud Saimpont & Angela Sirigu
Institut des sciences cognitives, Lyon (France)
Memory for Actions: Facilitating Role of Semantic Congruence

The enactment effect is a well known memory effect defined by an enhanced memory span for action sentences describing previously self-performed tasks as compared to heard-only sentences. One interpretation of the effect (Engelkamp & Zimmer 1985, 1989) suggests that subjects benefit from a multimodal encoding in which movement plays a major role. The existence of a motor contribution within the modality-specific memory subsystems is thus invoked. Neuroimaging studies (Nilsson et al 2000; Nyberg et al 2001) show an increase of activation in motor areas when subjects recall enacted or imagined phrases compared to purely verbally encoded ones. Interestingly, facilitation for verbal material seems to occur also when movements are not a direct enactment of the verbal material (Noice and Noice 1999, 2000), although at present this latter result has been demonstrated in professional actors only. It has been suggested that in this specific case the enactment effect resulted from a newly established association (i.e. for the purpose of the play) between speech and unrelated actions.
In this study, we directly address the issue whether the enactment effect can be found whenever a motor component is added to verbal encoding. Three experiments in which healthy right-handed subjects learnt a 30-items list were run. Each item in the list was an action phrase (auditory presentation) accompanied by a video-clip depicting the action. In Experiment 1 (congruent condition), eight subjects either observed or additionally acted out the action according to whether a green or a red cue appeared at the beginning of the video-clip. After 10-min interval (during which subjects were engaged in a visuo-spatial task), subjects were asked to freely recall and then to recognise among distractors all the phrases they heard. In Experiment 2 (motor incongruent condition), a new sample of eight subjects attended to the phrase/video-clip presentation but whenever the green cue appeared, they executed a standard unrelated movement (pulling a rope). Finally, in Experiment 3 (verbal incongruent condition), a new sample of eight subjects listened to an unrelated/non-action phrase while observing the video-clip and executing the corresponding movement in response to the green cue.
As expected, a clear enactment effect was found in the congruent condition (Experiment 1) where proportion of correctly remembered items was significantly higher for acted out phrases compared to viewed/listened ones in both free recall and recognition task. Response times were consistently faster for acted out items. On the contrary, no enactment effect was observed in Experiment 2 and 3, suggesting that this effect does not rely on an aspecific motor-induced facilitation, but rather requires a stable semantic correspondence between the phrase and the motor act. These results support the reactivation hypothesis in memory (Damasio 1989), and the existence of a specific motor memory subsystem.
 
 

Richard Dietz
University of Oxford (UK)
Is there such a thing as a vagueness-related conceptual role?

Starting from a standard propositional logic for languages with indefinite (or vague) sentences, we are faced with the following puzzle: Whether a sentence p of our language is indefinite or definite, there is either a positive or a negative answer to the question of whether it is the case that p or whether it is the case that not-p. However, if a sentence is indefinite, it is intuitively misguided to give one’s thought to this very question. In a recent paper, Hartry Field presents a conceptual role account of indefiniteness, which (according to him) provides a solution to this puzzle: If we consider a sentence as indefinite (on Field’s view) we simply have degrees of belief in the sentence and in its negation which deviate in some characteristic way from classical probability distributions. And this distinctive feature of supposedly indefinite sentences (according to Field) explains why we take speculations about the truth-value of a sentence as misguided, if we regard the sentence as indefinite. His account is in two regards radical. First it questions the widely held intuitive idea that if a sentence is regarded as indefinite, this amounts to an attitude to an ascription of indefiniteness to the sentence. And secondly, it challenges the classical model of how degrees of belief are structured. My arguments highlight some serious problems Field’s account is faced with.
 
 

Martin J. Doherty, Manuel Sprung, Nicolas McGuigan
University of Stirling (UK)
Here’s looking at you kid: Children’s understanding of eye-direction

This paper challenges the common assumption that children understand other’s visual attention from infancy.  Previous research is consistent with but does not require this conclusion.  In Experiment 1 we showed that children cannot judge eye-direction from schematic drawings until they are at least 3 years old.  In Experiment 2 we replicated this finding and extended it to photographic and real-life gaze stimuli.  Experiment 3 showed that even 2-year-old children can judge gaze from head direction but not eye-direction.  Judgement of eye-direction was associated with false belief understanding (r = .40, p < .05) although not when age was partialled out.  Experiment 4 compared our gaze judgement tasks to a surprising finding of Flavell, Shipstead & Croft (1978) in which 2-year-olds could hide an object by placing it behind a screen but not by placing a screen in front of it.  We found that the ability to hide objects by occlusion was strongly related to the ability to judge eye-direction.  This suggests that 2-year-olds understand attention in terms of general involvement in activities.  This concept would not allow distinction between being engaged with an object and being able to see it.  We conclude that children do not pay attention to eye-direction because they are not motivated to make this distinction until they acquire a more adult-like understanding of attention, probably part of more general theory of mind developments during the fourth year.
 
 

Stephan Dorsch
University College London (UK)
The unity of imagining

Brian O'Shaughnessy's recently published book "Consciousness and the World" is very special in many respects.  One less obvious of them is that it contains a unified account of the nature of imaginings, capturing both sensory and intellectual cases.  His envisaged theory is meant to provide not only a definition of the concept, but also an explanation of how the different kinds of imaginings do indeed satisfy this definition in virtue of certain properties that they instantiate. The main idea of his definition is that imaginings are similar to cognitive states, such as perceptions and states of knowledge, but necessarily never amount to them.
In what follows, I want to discuss whether his account is successful in achieving its aims, primarily by concentrating on his account of the relationship between imagination and cognition. It will turn out that I am often more in disagreement with the conclusions that O'Shaughnessy draws than with his explicit argumentation. One reasons for this is that he often does not spell out, or only suggests, what his arguments are. Hence, a substantial part of the essay will consist in interpreting and adding to these arguments (in conformity with what he says as well as what I take to be plausible) and hence in scrutinizing a position that is perhaps more O'Shaughnessian than O'Shaughnessy's. In other words, I will sometimes (especially later on in the discussion) use arguments that are more or less ad hominem. Besides, I will ignore the fact that O'Shaughnessy believes in aspect dualism as well as the sense-data theory of perception, since both do not significantly bear on the issue under discussion.
The first part of the essay will consist of the presentation of both O'Shaughnessy's (partial) definition of imagining and his two arguments in support of it; in the second, I will then in turn consider various objections against the two arguments and, in the course of the discussion, begin to sketch an alternative account of imagining in terms of will-susceptibility, rather than cognition.
 
 
 

James A. Hampton and Iben Kempff-Andersen
City University of London (UK)
Category-based Induction  - an effect of conclusion typicality

We considered the problem of category-based induction – the willingness of a thinker to project some newly learned property of one category member to other members of the same category.  Previous research has established that the perceived strength of arguments of the form “Category member A has property P, therefore category member B has property P” is influenced by the similarity of A to B, and by the typicality or representativeness of A in the shared category.  (The nature of P is also crucial, but we do not examine it in this study).   There is however little evidence that the relation between B and the category is influential.  We present two experiments that were designed to test whether the typicality of B also has an effect on inductive argument strength.  Using a regression (study 1) and an experimental (study 2) design, we show an effect of conclusion typicality such that people are more willing to project properties to more typical conclusions.  Our results are interpreted in terms of an associative model.
 
 

Christophe Heintz
Institut Jean-Nicod, Paris (France)

Word meanings and the Epidemiology of representations

 The problem of induction - the fact that a finite number of cases can never guarantee the validity of a general law - applies to lexical semantics. Its semantic version asserts that the meaning of a word cannot be wholly determined by the (finite) pasts uses of that word. The argument is logical. It has some strong consequences on epistemology. For theories of communication, however, it is somewhat of a puzzle, for people use words as if they had a definite determined meaning, while pasts use of the words remain finite. The finiteness is present at the level of the individual, whose learning of words meaning is based on a finite number of ostensions, and at the level of the language community.
To illustrate this puzzle, philosophers have invented monster meanings such as grue, which means green before time t, blue afterwards (N. Goodman, 1955) and quad - which is such that x?y = x + y iff x & y < 68 and x?y = 5 otherwise (S. Kripke, 1982). There is no logical reason that guaranties that the meaning of green is not the one of grue, or that when one says 'plus' he does not mean quus. There are two straightforward answers to this semantic puzzle. The first one is to assert that, since past uses cannot determine the meaning of words, then this is determined by something else, viz. cognitive innate constraints. The second one asserts that meanings are not determined, and thus relative to individuals and always changing. The first answer does not take the induction problem seriously enough and can hardly account for mistakes, misunderstanding and cultural differences in the usage of words. The second answer renders fruitful communication mysterious.
In this poster, I will briefly review and criticise the solutions that have been given to the problem of induction in semantics. I will then introduce the notion of stabilisation as a fruitful, unifying notion for dealing empirically with the problem. The notion of stabilisation that I want to develop is borrowed from Dan Sperber's epidemiology of representation. Sperber (1996) develops a theory of the spread of representations among a community that allows both cultural and cognitive factors to be taken into account. In this framework, the problem of induction in semantics is tantamount to the fact that there is no reproduction procedure that allows a representation associated with a word in the mind of an individual to be identically reproduced in the mind of another individual. Therefore, the empirical question to be dealt with is the following: 'how do people manage to have representations associated with words that are sufficiently similar one to the other so as to allow fruitful communication to happen in most instances?' In other words, 'how do word meanings stabilise within a language community and, consequently, for the individuals of that community?'. I will attempt to show that stabilisation is a strong enough notion that allows directing empirical investigations for fruitful explanations.
 
 

Sergey G. Kasvinov
Department of Pre-school Education, ASB, Kharkov (Ukraine)

Periodic Table of Children’s Psyche Development Stages

The initial hypothesis of this theoretical investigation based on experimental data and generalizations accumulated in psychology is the assumption that the development of children’s psyche proceeds in accordance with the philosophic idea of spiral development, i.e. that features of the developing psyche repeat periodically, although not identically, for that would turn spiraling into movement in a circle; therefore, the system of children’s psyche development stages resembles the periodic table, where properties of chemical elements recur within a subgroup after each period in a slightly different form. It is obvious that exposure of such a regularity in psychology presents philosophical, scientific and practical interest.  And the development of children’s psyche indeed shows periodicity (“spirality”).
 
 

David Kelly
University of Durham (UK)
Mental translations

Until recently, it has largely been taken for granted that meaning could be transferred from one language to another, or rather reproduced in another language. The possibility of transferring a pattern (be it mental, semantic, phonetic, textual…) from one means of representation to another is not an obvious one. This paper proposes to link some of the contemporary cognitive and philosophical concerns with language, by means of translation. We shall start from the views of translation which appear to have underestimated its breadth and scope in human understanding, in view of drawing broader conclusions about its philosophical status.
This will be considered in three sections, each dealing with ‘approaches’ to translation, revolving around definitions of the term itself. The first section deals with Quine’s  ‘radical translation’ and the link with ‘relevance theory’ as presented in Sperber and Wilson.  These accounts revolve around a ‘faithfulness’ definition of the translation process.
Next, the revolutionary approach of Noam Chomsky will be viewed as an account of some underlying patterns of language-use and their implications for any theory of mind. This section will be focused upon the ‘Minimalist Program’  and other research which pursues the foundations laid out by Universal Linguistics signals that empirical investigations into mental architectures as signified in language. The definition of translation used to characterise the Chomskian approach will be ‘transfomational’ i.e. in terms of mental ‘transformations’.
Finally, a wider definition of translation as explaining the fundaments of the mental framework will be developed in the light of the ‘Translational Control’ which is currently in use in genetics.
As an area of inquiry in which empirical enquiry into formal language-behaviour has been successful, translation provides a bridge between the strengths of the philosophical and the psychological approaches to language.
 
 

Daniela Kloo & Josef Perner
Department of psychology, University of Salzburg (Austria)

Training theory of mind and executive functions

“Theory of mind” denotes the research area that investigates children's acquisition of the conceptual system underlying our everyday psychology, i.e., what people want, believe, feel, know, etc. An important step in the development of a theory of mind takes place around the age of 4 years: Children understand that one can be mistaken about the world, that is, they understand false beliefs.
At about the same age children improve markedly in self control. They master tasks requiring executive control like the Dimensional Change Card Sorting task (DCCS task). In this task children are told to sort cards first according to one dimension (e.g. colour) and then according to another dimension (e.g. shape). It is not until the age of 4 to 5 years that children continue to sort correctly after the sorting criterion has switched to the other dimension.
Several recent studies point to a developmental relation between the acquisition of a theory of mind and executive functions in the age range of 3 to 5 years. Five competing theoretical accounts try to explain this developmental link. Two of these theories envisage a functional dependency between the development of executive control and the acquisition of a theory of mind:
(1) Some authors suggest that theory of mind is a prerequisite for executive functions.
(2) Other researchers argue that executive control is a necessary requirement or building a theory of mind.
To explore the nature of the interdependence between theory of mind and executive functions we conducted a training study. One group of children received training with respect to the theory of mind (on the basis of the false-belief task) and another group was trained on executive functions (using the DCCS). A control group was trained either on number conservation tasks or on relative clauses. Training was conducted in 2 sessions over the course of 2 weeks. In each training group specific tasks were given and feedback was provided.
The results indicate that the false belief training generalises to the DCCS task and that the card sorting training improves false-belief performance. Therefore the present study cannot differentiate between the two competing functional theories, but it demonstrates the close relationship between theory of mind and executive functions. At least it supports the connection between the false-belief task and the DCCS.
 
 

B. Kotchoubey, A. Kuebler, N. Birbaumer
Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen (Germany)
Conscious perception of one's own brain processes

From the prevailing point of view, conscious perception emerges as a result of complex processing of raw information obtained from sense organs. An alternative account relates the emergence of conscious perception to action and control, rather than to sense data. From this point of view we are able to perceive a variable if we are able to control it. It is known since the late 50th that people can learn to control the electrical activity of their brain (the EEG), but it remains unknown whether they can perceive these electrical processes. Note that the brain tissue, in contrast to most visceral organs, contains no specialized receptors that could deliver information of its functional state. In the main experiment twenty-two epilepsy patients learned to control low-frequency components of their EEG during 35 training sessions. In particular test sessions they were asked to estimate the actual state of the controlled function. After 30 sessions they were able to do this significantly above the chance level. This ability to estimate the EEG process apeeared much later in the course of training than the ability to control this process (after 10-15 training sessions).This is in line with the hypothesis that conscious perception emerges on the basis of control, but at odds with the idea that adequate perception is the necessary prerequisite for reliable control. None of other physiological variables measured in that experiment correlated with the patients' subjective ratings, which indicates that their estimations may be specifically related to that only variable they learned to control. In the control experiment a similar procedure was conducted with a patient with amyotrophic lateral sclerosis who was almost completely paralyzed, artificially ventilated and fed. He was able to express his subjective estimations using an artificial larynx cannula. Also this patient was able to highly-significant above-chance estimation of the EEG process he had learnt to control, suggesting that peripheral muscular activity is not necessary for this perception. To sum, the data confirm the hypothesis that reliable control over a physiological function can give rise to a new kind of perception notwithstanding the lack of specialized receptors delivering "raw information".
 
 

Dorothée Legrand
Ceperc – Department of philosophy – University of Provence (France)
Me, I, Myself

The self is a central notion both for psychologists and philosophers but none of these disciplines offer an univocal definition of it. I propose here to clarify this notion by a triple analyse of it, distinguishing between "ME", "I" and "MYSELF". First, "ME" is the person that I and others (at least in part) can describe as being me. To know "ME" is to know the diverse properties I possess. At this level, I am conscious of who I am. Second, "I" is what I refer to each time I say "I", or each time I could have expressed my experience with this term, even if I do not actually express it. At this level, I am conscious that I am. Third, "MYSELF" is the subject to whom something appears somehow in each phenomenal experience. This subject can be transparent to himself, not being conscious of himself having this experience. At this level, I am, without being conscious of it. This paper details these three levels of self. Moreover, it shows that "ME" is not enough in itself to define the self, because it does not assure me that I am the one who I am conscious of. "I", on the contrary, is immune to error through misidentification. In this sense, "I" is a necessary level of self which found "ME". However, "I" is not the most basic level of self. Indeed, "I" is the conscious subject of conscious experience. This subject is still present even when not conscious of himself. Thus, "MYSELF", as it is an unconscious "I", is also a level of self. We end up with a triple notion of self, where no level is sufficient, but all three are necessary to describe completely what the human self is.
 
 

Huib Looren de Jong
Vrije Universiteit, Amsterdam (The Netherlands)
Reduction, elimination and the dynamics of explanation

In this paper, some problems with a recent proposal for psychoneural reduction, New Wave Reductionism (Bickle, 1998, 2001, 2002) are discussed.
The unique selling point of NRW is the continuum ranging from retentive reduction to elimination, that replaces the classical bridge laws as the mainstay of reductive inter-level relations. This continuum is criticized as too low-dimensional for a realistic view of theory dynamics. It is seems limited to the diachronic dimension of theory succession, whereas synchronous coexistence of theories, mutually constraining each other, may be as interesting.
A close look at Bickle’s treatment of Multiple Realisability and his idea of “structuring by reduction” indicates that these are too simple. A more pluralist, non-reductive view of multiple realisation seems required. Furthermore, the notion of coevolution of theories (which Bickle subscribes to) seems to imply a more important role for higher level (functional) theory than can be accommodated by his reductionism. It is argued that an initial functional characterization may be presupposed in a reductive explanation, and that this constitutes an obstacle for reduction (Kincaid, 1990).
The conclusion is that by replacing the classical bridge laws with a continuum of theory correction NWR has made considerable progress beyond the classical D-N model in understanding explanatory intertheoretical relations. However, a more pluralist view on reduction is required. Coexistence, coevolution, and mutual selection pressure, both bottom-up and top-down, of higher and lower level theories rather than theory succession is a plausible or at least conceivable empirical possibility (McCauley, 1996).
 
 

Edouard Machery
Institut Jean-Nicod, Paris (France)
Should really philosophers rule over psychologists?What’s wrong with Peacocke’s view of the psychology of concepts

Peacocke’s theory of concepts implies a puzzling unequal division of labor between philosophers and psychologists. Philosophers have to characterize a priori the possession conditions of concepts, i.e., to answer to the question, What is it to possess the concept ? ? Psychologists on the contrary are supposed to describe the nature of the subpersonal states and processes required for someone to meet the possession conditions established independently by philosophers. Ultimately, this unfair division of labor rests on the alleged a priori nature of the individuation of concepts and on the supposed capacity of the philosophical methodology to provide such an a priori individuation. This paper focuses on Peacocke’s version of this methodology. On the basis of the principle of Cognitive Plausibility, according to which any adequate reasoning methodology should be consistent with our actual way of reasoning, this paper maintains that Peacocke’s methodology is inadequate, for it isn’t apt to provide any knowledge at all. It hangs on a an empirical claim that is not well supported by the current evidence: human beings’ counterfactual judgments are based on the off-line simulation of actual judgments.
 
 
 

Jean-Philippe Magué
Dynamique du langage, Université Lyon 2 (France)

Emergence in a population of agents of a lexicon based on an individual conceptualization

This work deals with the emergence of a lexicon based on an individual conceptualization of the world in a population of agents. The multi-agents paradigm seems well suited for the investigation of the emergence of language. Previous works have shown that, language games, i.e. simple well-defined, communicative interactions, permit populations of agents to develop shared realistic phonological systems, elements of syntax or lexicon. In most previous attempts (Steels, 2000), words used by agents were to name single object of the world, single agent or single feature. The conceptual structure presented here is richer; we can name an object a “hammer”, a “tool” or an “artifact” according to our purpose. In this study, agents construct an individual representation of world that they use to communicate not only about the objects of the world, but also about categories of objects. Conceptual spaces have been proposed by Gärdenfors (2000) to be a framework for modeling the formation and the evolution of concepts. They are multi-dimensional geometric spaces. The different dimensions, named quality dimension, correspond to the different ways objects can be judged similar. Within the conceptual space, particular points are the center of spheres that represent prototype of categories. Classical realistic semantics states that meaning is a mapping from a word into an object or a relation between objects of the world. In contrast, cognitive semantics argue that words are mapped into mental entities. Gärdenfors propose that concepts as modeled in conceptual spaces are good candidates for these mental entities. On the basis of these considerations, we constructed a world composed of objects and agents endowed with conceptual spaces, initially empty, and a lexicon that associates each prototype with one or more words (associations are weighted with a score). Agents play a modified version of a language game known as the naming game. In the game they play, an agent, the initiator, chooses a set of objects, and communicates his choice to another agent, the recipient, in a non linguistic way. Then the recipient categorizes the objects according to his representation of the world by determining which prototype is the closest of the gravity center of the projections of the objects in his conceptual space. The concept represented by the prototype is more or less abstract, depending on the similarity of the objects. The recipient indicates to the initiator how he calls the concept, and the initiator, determines whether he agrees with the word user by the recipient. Depending of the issue of the game, both agents modify their lexicon and/or their conceptual space: they may modify the score of the associations, acquire a new word or a new concept, or modify the structure of their conceptual space. The results of the simulations show that although agents develop their own representation of the world in their conceptual space, a shared lexicon emerge from the population and allows the agents to communicate not only about the objects, but also about categories that regroup objects and/or other categories.

References:
Gärdenfors, P. (2000), Conceptual Spaces: The Geometry of Thought, Cambridge: The MIT Press.
Steels, L. (2000), Language as a Complex Adaptive System. In M. Schoenauer (Ed.), Proceedings of Parallel Problem Solving from Nature VI, Berlin, Germany : Springer-Verlag.
 

Genoveva Martí
London School of Economics (UK)
Substitution and Identity

One of the many important tasks of semantics is to provide an account of the substitution patterns of a language, i.e., to furnish an explanation of the conditions under which semantic values of complexes are preserved when components are replaced. This is why it is important to articulate an acceptable principle of substitution. A traditional statement of the principle of substitution holds that co-designative terms should be intersubtitutable salva veritate. In the past I have argued that such a principle is not connected properly to some pre-theoretical data about the function of referring terms. In this paper I argue that such a principle does not gain support either from other, more fundamental, laws. I end up defending the adequacy of a different principle of substitution.
 
 

Albert Newen, Rimas ?uplinskas
Philosophisches Seminar LBFIII, University of Bonn (Germany)
Mental Causation: A real phenomenon in a physicalistic world without epiphenomenalism or overdetermination

The so-called problem of mental causation as discussed in the recent literature raises three central challenges for an adequate solution from a physicalist perspective: the threat of epiphenomenalism, the problem of externalism (or the difficulty in accounting for the causal efficacy of extrinsic mental properties) and the problem of causal exclusion (or the threat of overdetermination). We wish to account for mental causation as a real phenomenon within a physicalistic framework without accepting epiphenomenalism or overdetermination. The key ideas of our proposal are an internal realism of causation combined with a relative notion of individuating events. We are arguing – contra Davidson – that there is no absolute notion of events (neither as types nor as tokens) but rather one which is relative to explanatory interests and our intuitions concerning a relevant spatial and temporal overlap. Furthermore, we are presupposing a metaphysics of internal realism: We can only characterize entities by means of concepts produced within our epistemological framework. Physical concepts and mental concepts crossclassify the world as it is. Relying on this framework we try to explain how mental causation can be adequately described: Although mental concepts are not reducible to physical concepts and mental event-tokens may be different from “underlying” physical event-tokens, mental events are real phenomena that are realized by physical phenomena in special context-conditions.
 
 

Anne Newstead
Wolfson College, Oxford (UK)
Agency, Perception, and Self-Awareness

 In my paper, I seek to revive, repair, and in some cases, supply the arguments for GEM Anscombe’s claim in Intention that agents have a special ‘non-observational’ awareness of their intentional actions. Developmental psychologists and philosophers alike recognise that awareness of one’s intentions and actions is an important and fairly primitive (developmentally basic) form of self-awareness. There has been some debate recently as to whether self-awareness can be modelled on a subject’s perception of an object in the external world. I argue, by bringing some of Anscombe’s ideas into a contemporary setting, that in the central case of awareness of one’s own agency, the perceptual model does not work. An agent’s awareness of her intentional action is different from her perceptual awareness of an external object in the world, even though in both cases a direct theory (excluding intermediaries between such awareness and its object) is appropriate.  In the case of (conceptual) perceptual awareness of a scene, the content of one’s awareness is a propositional content that is satisfied by and causally dependent on the way the world is.  In the case of agent’s awareness, the content of that awareness is an intention that is satisfied when the agent acts in accordance with it, and is such that it exerts a causal influence on the way the agent (tries to) act.  I argue that proprioceptive and kinaesthetic feedback fall into the category of perceptual awareness, because the direction of fit and causation of such awareness is, as with all perceptual awareness, from world to mind. Such sensory feedback is of the wrong sort to support an agent’s awareness and cannot, in any case, explain the anticipatory aspect of an agent’s awareness.
 I endorse the intuition that an agent cannot relate to her own actions in the way that she relates perceptually to an external object, being careful to differentiate this claim from the implausibly strong claim that one cannot perceive what one is doing.  I then claim that it is constitutive of full-blooded, self-conscious agents that they be able to relate to their actions in this non-perceptual mode of awareness. I illustrate this claim by a consideration of Anscombe’s thought experiment of the ‘A’ users, a community of language users that uses ‘A’ as a device of self-reference, but ascribes ‘A’ on a purely perceptual basis.  I argue, along with McDowell, that we cannot really make sense of such beings as agents.  I conclude that agency requires self-awareness, where this self-awareness is of the non-perceptual kind.  However, since such agency is itself a form of self-awareness, this should come as no surprise.
 

Analucy Aury Vieira de Oliveira
Department of Foreign Languages and Translation, University of Brasilia (Brazil)
Psycholinguistics in second language learning

 (INTRODUCTION): The research project Psycholinguistics in Language Learning became true because of the doubts of several language teachers in Brazil. This work aims to find out what are the psychological, linguistics and pedagogical problems  that don’t allow the  brazilian students to realize an appropriate learning of a second language, with this investigation we might develop or choose a better methodology to improve the process of learning of our University students. (METHODOLOGY): Questions, interviews and observations based on the the theories of famous Linguists such as  Noam Chomsky and Stephen Krashen; and the  observation of the language studentes from the  1st. to 8 th. periods of the Letters course of  CEULP ( Lutheran University Center in Brazil). (RESULTS AND CONCLUSIONS) The results seems to prove the hypothesis of  Krashen (1985), “that Acquisition  is a subconcious process identical in all important ways to the process children utilize in acquiring their first language, while learning is a concious process that results in knowing about language”. The hypothesis affective filter, also from Krashen (1985) “ The affective filter is a mental block that prevents acquirers from fully utilizing the comprehensible input they receive for languages acquisition”, is another important aspect that influences in Language Learning.(CONCLUSION)  Psychologycal and afective problems may cause a direct impactic in learnig such as : demotivation, perfeccionism, low self stem,  excessive awareness  and nationalism. All of these facts are the results of the previous life of the students may occur in adolescents, as well as in adults.
 
 

Ahti-Veikko Pietarinen
University of Helsinki (Finland)

Logic, language-games and ludics : some wittgensteinian themes in the foundations of logic and computation
 

Wittgenstein's language-games can be put into a wider service by looking at their contribution to contemporary notions of logic and semantics of computation. I will highlight two examples : the role of language-games in recent logical studies, and in some recent advancements in computer science. The paradigm of computation termed `ludics', for instance, has a striking proximity to language-games. The interrelations that are emerging can be scrutinised from the viewpoint of logic that virtually necessitates game-theoretic conceptualisations. These conceptualisations in turn are related to semantic issues on language and demonstrate that meaning of utterances can in at least many situations understood as Wittgensteinian language-games of `showing or telling what one sees'.
 
 

Manuel de Pinedo Garcia
Department of Philosophy, University of Grenada (Spain)

Computation and Mind: From Syntax to Pragmatics

Two interrelated themes—syntactic theories of mind and understandings of intelligence which focus on explicit, representational, knowledge—have come increasingly under fire. In recent debates much emphasis has been put on the practical aspects of knowledge as an alternative to those themes. This short paper deals with the conflict between both approaches to the mind in the field of artificial intelligence, and briefly surveys several areas where that conflict has been felt. I will point out difficulties within a specific field of much contemporary relevance (the theory of computation and its relation to philosophy of mind) which are connected to well-known problems of representationalism, i.e. the doctrine committed to the possibility of individuating intentional and semantic items through non-intentional and non-semantic means. My purpose here is not to embrace a particular scientific theory of the mind over another. I do not believe that research on the enabling conditions of mentality amounts to a theory of the mind. But I want to suggest that even research concerned with subpersonal states wins in productivity if it does not ignore that thought is constitutively corporal and intersubjective.
 I will review Dreyfus’ criticism of representational theories of mind which he connects to traditional Artificial Intelligence and to Husserl’s intellectualist phenomenology. Parallel to this criticism is Brian Smith’s insistence that a theory of computation should give priority to semantics with respect to syntax, and his reaction against computation understood as explicit-rule following. I will briefly look at his work and relate it to the idea that logic should be thought of as an area of pragmatics and not as a syntactic study. I will finish by looking at a more recent paradigm in theories of computation, Artificial Life, which does not share the theoretical commitments of traditional AI, and has proved more promising in areas where the latter was at a standstill.
 
 

Andres Posada
Institut des sciences cognitives, Lyon (France)
The theory of layers

The theory of layers is an analysis based on two points of view regarding the human consciousness. The first of these is known as the empirical, or third person view. It  refers to the rational analysis of the components of the world perceived by the senses and is the view taken by traditional science when analyzing the world. This point of view proposes that the brain is a necessary interface between the external world and the conscious perception of the world by the individual. There is several steps between an object in the world and its conscious perception: the physical proprieties of the object in the environment are detected by the sensory organs, an electrical code is then sent to the brain, the brain processes the information and then the object enters into the consciousness of the individual.
The second point of view is the first person perspective, which describes the mechanisms and components of our internal world of consciousness. It is used by philosophers to explore and understand what makes us be what we are. The internal consciousness can be usefully divided into two components: Qualia and Think. Qualia are references to the phenomenological character of our experience, to the way in which they appear to the conscious subject – the redness of red, or the tactile sensation of crushed velvet –. The Think refers to ideas, concepts, mechanism of reasoning, etc. It is not only the dynamic processes involved in thinking, but also those attributes of thought which are traditionally seen as static such as the representation of objects or concepts.
Combining both of these perspectives, an interesting paradox appears about what we are conscious of and how we experience it. The reality for an individual is its own world of consciousness and this is strongly bound to the detectable physical proprieties of the external world. However, since Qualia and concepts are components of our world of consciousness but they are not the external world, how the outer world is like? A gap clearly exists between both inner and outer worlds of consciousness; it depends on the detection range of the sensory organs end their extensions (ex: microscope, radio). However the limits of detection can move and can qualia become detectable when this limit is moved?
 
 

Nausicaa Pouscoulous, Ira Noveck, Guy Politzer, Anne Bastide
Institut des sciences cognitives, Lyon (France)
Evidence for Scalar Implicatures in Young Children

Implicatures are a set of propositions a hearer can infer in context from an utterance, propositions that go beyond the explicit, literal meaning of the sentence. We focus on a paradigmatic type of implicature : scalar implicatures. These occur when a speaker chooses to articulate a weaker word on a lexical scale of informativeness in such a way that the hearer believes that the speaker intended not to use a stronger one - either because she didn’t have sufficient information or because she knew that it was inappropriate to use the stronger one. For example, Some is a weaker quantifier than All and an utterance like Some children have teddy bears can be taken to imply that Not all children have teddy bears. We present two developmental experiments investigating when children use scalar implicatures. These studies follow up on work carried out by Noveck (2001), who showed that children do not draw on implicatures as readily as adults, and tend to interpret a weaker word as compatible with its stronger equivalent (e.g., interpreting some to be consistent with all). Experiment 1 generalised the findings from Noveck by using a novel methodology to demonstrate the same trend. Experiment 2 demonstrated that even very young children (e.g. 4 year olds) can make implicatures provided that the cognitive effort required is sufficiently low. We discuss the results of our experiments in terms of Relevance Theory (e.g. Sperber & Wilson, 1986/95).
 

Panu Raatikainen
University of Helsinski (Finland)
Thinking what’s not there

One key puzzle in the theory of mind, often called the problem of  intentionality, is our undeniable ability to have thoughts about  objects that do not exists. How is this possible? How should one understand the objects of such thoughts?
The Meinongian approach, which postulates non-existing entities  for this purpose, is arguably implausible. A much better solution  was proposed, in the context of names,  by Russell. He argued  that names are in fact abbreviations of certain definite descriptions, so that if nothing actually satisfies the description, the name is empty. The same approach can be easily applied also in the level of thoughts.  But elegant as this solution is, it unfortunately faces difficult problems of its own. Kripke and Donnellan have argued convincingly against the view that names are synonymous with some descriptions. Against this, they have suggested that at least usually, a name refers in virtue of some factual, external causal- historical relations between users of the name and the object named. Just as Russelian approach, this approach is naturally applicable also to thoughts. Such a causal-historical approach, also called “externalism”, is quite popular nowadays.
One has complained, however, that the causal-historical approach is unable to deal with the case where one’s talk or thoughts are about something that does not exist.  My aim is to give a systematic reply to this worry. I begin by proposing that that one should draw a distinction at the level of thoughts and concepts, based on Donnellan’s distinction between attributive and referential uses of descriptions. That is, I suggest that thoughts can be thoughts of something in two rather different ways. One may have e.g. the thought “The murder of Smith is insane” with no particular person in mind, i.e. about whoever committed the murder. This is the attributive case. But one can have the same thought also with a particular person in mind (e.g. in perceptual contact) - even if the person in question is in fact innocent. In this case, the thought relates to its object referentially.
Such a distinction is rather plausible in itself, but it has an additional welcome consequence that it provides a natural way for a causal-historical theory to handle the problem of non-existing objects of thought.  The idea is that only those thoughts that function referentially get their referents in the way the causal- historical approach describes the issue. There are other sort of thoughts which function attributively and are description-like. I argue that typically thoughts whose objects do not exist belong to this latter category. They aim to pick their referents via some description, and if nothing in fact satisfies the description, they are non-referring. Thus my solution is, after all, essentially Russellian, but it avoids the problems of Russellian general theory of names. It assumes that in many cases, the causal-historical account is the correct one. Arguably this is all that the causal-historical theory has ever claimed to be doing.
 

Joanna Raczaszek
University of Warsaw (Poland)
Symbolic and dynamic aspects of language

The present paper is an attempt to look at natural language as a dynamic phenomenon. What is traditionally considered a language, that is “a static system of symbols” is seen as just an element of a larger system. We try to identify its dynamic elements and establish their role with respect to the “system of symbols”. We focus on three dynamical aspects, each taking place on a different time-scale: evolutionary language change (i.e., a diachronic aspect), processes of communication (language use) and language acquisition. The relation between symbolic and dynamic in language is encompassed in the framework proposed by Howard Pattee to deal with information in biological systems. Symbols of language are information bearing entities that emerge under the pressures of communicative needs and serve as concrete constraints on development and communication. Including in the picture of language a variety of dynamic processes and acknowledging their role in shaping and sustaining the structures of natural language sheds new light on the nature of symbols themselves, issues as difficult as characterization of linguistic meaning, and the effects of context. It also calls for a change in the methodological frameworks for studying language. Broadening the range of phenomena relevant for the study of language makes it clear that a synchronic analysis of a system of meaning-containing entities is not sufficient to find answers to some problems recurring in linguistics and philosophy of language. A more convenient research framework may include methods designed specifically for studying complex dynamical systems, systems in which processes of self-organization take place over different time scales.

Key terms: language, symbol, meaning, context, dynamical systems
 

Athanassios Raftopoulos and Vincent C. Müller
American College of Thessaloniki
A Step Toward Solving the Grounding Problem

In this paper we address the issue of grounding for experiential concepts. To avoid ‘encodingism’, that is, relating representations to representations, we postulate that the process of reference fixing must be bottom-up and non-conceptual, so that it can break the circle of conceptual content and touch the world. Given that perceptual demonstratives are a basic form of such concepts, we examine ways of fixing the referents of such demonstratives. For that purpose, an appropriate causal relation between representations and the world is needed. We claim that this relation is provided by spatial and object-centered attention that leads to the formation of object-files through the function of deictic acts. This entire causal process takes place at a pre-conceptual level, meeting the requirement for a solution to the grounding problem. Finally we claim that our account captures fundamental insights in Putnam’s and Kripke’s work on “new” reference.
 
 

Anne Reboul
Institut des sciences cognitives, Lyon (France)
A reductio ad absurdum argument against philosophical zombies

Chalmers' philosophical zombie argument is one of the major philosophical argument against functionalist reductionism, i.e. the idea that the phenomenal can be reduced to the physical via a functionalist bridge. It rests on the logical possibility of philosophical zombies, that is creatures microphysically and functionally identical with human beings, but deprived of qualia.If Chalmers is right, this means that neurosciences will never be able, no matter how sophisticated they can get, to approach consciousness. However, it is not clear that the PHZombie argument is correct and I think that it can fall under a reductio ad absurdum.
Functional equivalence entails behavioral identity. Hence, any act of an Earthling (let's call her Hannah) will be identical with the corresponding act of her PHZearth counterpart (let's call her PHZhannah). Notably, their speech acts will be identical. Thus, when Hannah says: "I'm conscious", her locutionary and illocutionary acts are identical with those of PHZhannah, who is engaged at the same time in saying: "I'm conscious". However, though this is true of their locutionary acts, where syntactic identity suffices to ensure identity, it is clearly not the case that their illocutionary acts are identical. This is because, though "conscious" will pick up what it is like to be Hannah, it cannot pick up anything of the kind regarding PHZhannah because there is nothing it is like to be PHZhannah. Hence, Hannah and PHZhannah are not behaviorally identical: some of their illocutionary acts, those concerning qualia and phenomenal consciousness, will not be identical. If this is the case, Hannah and PHZhannah are not functionally identical.
That's the reductio ad absurdum and it is my contention that this undermines Chalmers' argument.
 
 

Paula Rubio Fernández
Research centre for english and applied linguistics, Cambridge University (UK)

Understanding nominal metaphors: beyond class inclusion.

According to the class-inclusion model of metaphor interpretation (Glucksberg & Keysar, 1990), a nominal metaphor such as My job is a jail would be understood as a categorisation statement where jail would not only denote what it usually does, but broader an ad hoc category including both jails and my job. Jail, as the metaphor vehicle, would now give name to the new category while its salient properties would become prototypical of that category. Since my job, as the metaphor topic, also denotes a member of the ad hoc category, typical properties of jails, such as being unpleasant, constraining and punishing would now be attributed to it.
Despite its explanatory power, the class-inclusion model has certain inconsistencies, which cast serious doubts on the validity of the model and the interpretability of their experimental results. First, the same arguments that Glucksberg and Keysar give to demonstrate that similes are implicit metaphors show that nominal metaphors are not necessarily interpreted as categorisation statements. Second, the claim that metaphor interpretation is based on the prototypicality of the metaphor vehicle is too constrained to account for cases of figurative language use where interpretation may be based on a peripheral albeit relevant aspect of the vehicle. Third, the assumption that the literal referent of the metaphor vehicle is also included in the new category poses certain problems for the construction of the ad hoc category, as well as for the interpretation of their experimental results.
In this paper, I propose a defective frame approach to metaphor interpretation, which adapts a conceptual frame account of human knowledge structure to the relevance theoretic model of lexical meaning (see Barsalou, 1992; Sperber & Wilson, 1997). Although also involving concept construction, the defective frame approach makes different claims than Glucksberg and Keysar's model about the ad hoc concepts constructed in metaphor interpretation and therefore different empirical predictions. We present the results of a cross-modal lexical priming experiment, which are predicted by the defective frame approach but not the class-inclusion model.
 
 

Isabel Saillot

Layer 5 Paleocognitive Study Of Scladina Cave, Sclayn, Belgium
 

In the field of paleolithic cognitive processes, most authors have defined their own vocabulary. No one seems interested in the use of the concepts developped since many years by the field of cognitive science, especially experimental psychology. Cognitive psychology of goal oriented actions seems nevertheless a powerful tool provider in the attempt to modelize paleolithic man behaviour. Here we show how cognitive sciences can help prehistorians to interpreted animal remains as a window on Paleolithic man cognition related to subsistence behaviour : Procedures applied to objects can be seen as properties of these objects exactly the same way than color or shape for instance : objects have functional properties as well as surface ones. This result is very interesting for the prehistorian because procedures are often known by deduction along with the artefacts directly found on the site. What we know about actions is structured in our memory like what we know about objects. It is hence possible to define a semantic action network as it is possible to define an object semantic network. The method for the prehistorian lies in analysing paleolithic men activities step by step, defining precisely the objects and the procedures they underwent by the men on the site. What paleolithic men did whith the objects found on the site is the field of palethnology, and the palethnologic results are the starting point for the paleocognitician.
The mammals processing chain of layer 5 in Scladina cave is analysed. The objects prehistoric man considered, and their properties, are the elements of the analysis. The objetcs and their properties are linked by binary relationships and then processed by SIMBOL, a software designed at the Laboratory, after an original publication by the Laboratoire de Psychologie Expérimentale of University Paris 8. SIMBOL returns a graph that allows a rigorous description of paleolithic categorization process which occured during the use of the site.
According to archeozoological and palethnological results, the cognitive categories of a neandertal man or population, regarding elementary tasks of the subsistance behaviour are displayed by SIMBOL. Cognitive complexity and fluidity seem in a medium range and will necessitate further investigations to deliver relevant information. But the structure of the graph already shows that that the three medium range categories can be seen as « tool » ones : the similarity of their location on the graph could go along with a cognitive similarity. The semantic analysis method should be applied on each prehistoric site, systematically associated to the digging process, like any other caracterization method regarding datation, climatic, faunal, geologic or lithic data. This method will deliver all its power when there are enough results to allow comparisons between sites of different periods (giving insights about evolution) or different places of the same period (giving insights about prehistoric culture).
 
 

Hans-Christian Schmitz
IKP, University of Bonn (Germany)

Accents and Information

Hypotheses: (1) Accents are used to emphasize constituents in spoken language. Due to acoustic disturbances and variations in attention, the recipient of a spoken utterance may perceive only parts of the utterance. Nevertheless, perceiving parts can be sufficient for understanding the whole. Emphasizing a constituent raises the probability of its being perceived. The less constituents are emphasized the higher the probability that the recipient will perceive all emphasized constituents. A speaker has to utter a sentence in a way that makes the recipient perceive at least those constituents which are necessary and sufficent for understanding the entire sentence. In spoken language the speaker can do so by accentuating. Optimal accentuating is accentuating the minimal set of constituents needed for comprehension. The set is determined by the context of the utterance, especially the question under discussion. (2) Although a speaker does not always accentuate optimally, it is commonly taken for granted that he does. Even if the recipient perceives only some constituents of an utterance, he usually does not have to ask the speaker to repeat. He can rely on having perceived the right constituents which lead him to proper understanding. If a recipient fully perceives an utterance, he can then conclude what the speaker takes as being the context of his utterance and -- in case the context determines the meaning -- what the speaker means by his utterance.
Out of context the sentence Peter only watches football is ambiguous. There are three possible readings: (i) Peter watches football and from a set of alternatives to football he does not watch anything else, e.g. tennis. (ii) Peter watches football and he does not stand in any alternative relation to football, e.g. he does not play football. (iii) Peter watches football and, from a set of alternative properties, he does not have any other property, e.g. doing the laundry. A speaker who accentuates optimally and who utters the sentence with accents on only and football takes the recipient to understand his utterance just by perceiving only and football . The recipient understands the sentence on the basis of this perception if he knows that the question under discussion is what Peter watches. If the recipient perceives the complete sentence, then he can conclude from the choice of accents that the speaker believes that the recipient believes that the question under discussion is what Peter wants. If the recipient also believes that the speaker gives relevant information in this context, then he concludes that the speaker means reading (i).
The disambiguating force of accents is also explained by focus theories (e.g. [1], [2]). Predictions on accentuation and interpretation based on the hypotheses above are in some cases different from the corresponding predictions made by focus theories. Predictions differ concerning the accentuation of focus operators (following focus theories, only does not have to be accentuated for any reading of the example above), focus projection (following focus theories, accentuation does not distinguish reading (i) and (iii)) and the determination of contextual alternative sets. The poster presents empirical evaluations on these differences.

References
[1] Krifka, Manfred (1996): Frameworks for the Representation of Focus, paper presented at ESSLI 1996, Prague.
[2] Rooth, Mats (1995): Focus, in: Lappin, S. (ed), Handbook of Contemporary Semantic Theory, Oxford, 271-297.
 
 

Dov Shmotkin
Tel-Aviv University (Israel)
The Pursuit of Happiness in a Hostile World: The Role of Subjective Well-Being Reconsidered

Subjective well-being (SWB) refers to people’s evaluations of their lives, and is usually conceived as happiness or satisfaction with life. This is an established research area that hardly relies on the philosophical traditions addressing the concept of happiness in ancient and more recent time. In contrast to the complex and the controversial philosophical discourse regarding happiness, empirical studies usually resorts to very simple, indeed simplistic, assessment of SWB. Nevertheless, the empirical research has persistently produced certain counterintuitive findings on SWB. According to these findings, objective life conditions have little effect on SWB; life events affect SWB only for a short period; aging does not necessarily involve a decline of SWB; positive affect and negative affect, which are major SWB dimensions, appear relatively independent; SWB is largely based on self-illusion; and, finally, most people are found happy. The fact that these findings go contrary to some prevalent expectations raises the need to reconsider the notion of happiness in a more dynamic conceptualization. In a model I have formulated, it is proposed that a primary role of SWB is to constitute a favorable psychological environment, which functions to counteract the ever-present hostile-world scenario (HWS). HWS is defined as one’s image of actual or potential threats to one’s life or integrity. This concept has deep roots in the human psyche. Both SWB and the HWS are presented as complementary adaptational systems, which employ various psychological mechanisms in order to regulate each other. There are functional limits of SWB: thus, SWB is often inefficient in coping with trauma, which often turns to be a chronically activated HWS; also, SWB does not substitute for meaning in life. The proposed model calls for a new evaluation of the philosophical discussions of happiness, dwelling on the dialectical manifestations of the pursuit of happiness in a hostile world.
 

M.K.D. Schouten & F. A. Keijzer
Vrije Universiteit, Amsterdam (The Netherlands)
Embedded Cognition and Wide Mental Causation

Kim’s Causal Exclusion Problem states that the supervenience of mind on matter  threatens to rob mental properties of all causal powers. It has been argued that this problem can be deflated: it must be wrong because if it works the causal efficacy of all macroproperties would be put in jeopardy. Kim’s has objected to this argument by pointing out that the difference between mental and neural properties is one between orders, not levels, and that the Causal Exclusion Problem only arises in such intralevel contexts. In this paper we argue against Kim’s presumption that mental properties supervene on brain properties. Recent findings and theorizing in the cognitive and neurosciences strongly suggest that mental properties are genuine macroproperties. That is, mental properties are micro-based and have new causal powers that go beyond the causal powers of the underlying microstructure.
 
 
 

Mark Siebel
University of Leipzig, Department of Logic and Philosophy of Science (Germany)
There’s something about Linda. Probability, coherence and rationality

In one of Tversky’s and Kahneman’s most famous experiments, subjects read a brief personality sketch, ‘Linda is single, outspoken, was deeply concerned with issues of discrimination …’, and have to tell whether it is more likely that Linda is a bankteller or a bankteller and active in the feminist movement. The result seems to show that even statistically sophisticated people are prone to violating the probabilistic principle that a conjunction cannot be more likely than one of its conjuncts. If the probability calculus provides the standard of rationality for that domain, it thus appears that their reasoning is not rational.
Gigerenzer, however, has pointed out that this conclusion is too hasty. Apart from the meaning layed down in the mathematical calculus, the word ‘probable’ has a number of non-mathematical interpretations. Arguably, Grice’s relevance maxim brings participants to understand the terms of reference in a way not intended by the experimenters because the story told about Linda is irrelevant to the intended task. And if they do not rank probabilities in the sense of probability theory, Gigerenzer continues, then they hardly violate a principle of that theory.
This raises the question of what kind of reasoning they perform instead. I suggest to account for subjects’ judgements by something which Thagard has successfully employed in other cognitive domains: they can be seen as resulting from inferences to the maximally coherent whole, where explanatory relations are a central factor in increasing coherence whereas (logical as well as weaker) inconsistencies diminish it. That is, participants choose the conjunction ‘Linda is a bankteller and is active in the feminist movement’ because various explanatory connections between the second conjunct and the personality sketch compensate for the inconsistencies to which the first conjunct gives rise. From a coherentist point of view, the conjunction is thus to be preferred over the simple hypothesis that Linda is a bankteller because it provides more coherence.
Finally, I examine the consequences of that account for people’s rationality. Note that a Subjective Bayesian will judge them to be irrational even if they do not assess probability but, e.g., coherence. For the result is anyhow that they place more confidence in a conjunction than in one of its parts. I argue, however, that to equate rationality with conformity to the probability calculus means to invoke a standard of rationality too narrow. Degrees of confidence conforming to the probability calculus are only one factor when it comes to the question of whether someone is a rational believer. There are further factors, and one of them is coherence. Briefly, one can do worse than choosing the most probable hypothesis; but often it is worth taking the risk of a low probability by selecting the assumption providing more coherence because this increases the chance of understanding.
 

Jonathan Smallwood
Caledonian University

Effort as an affordance : the consequences for our understanding of thinking
 

This paper is concerned with highlighting an inconsistency within the literature on task-unrelated thinking . This paper will demonstrate that in two related fields of cognitive research parallel yet contradictory assumptions are made regarding the consequences of the imposition of a cognitive load.   In the first area of research concerned with understanding the consequences of thought suppression (see Wegner, 2000 for a review) an imposition of a cognitive load increases the accessibility of the target thought (E.g. Wegner and Erber, 1992 Experiment Two).  In contrast, when researchers investigate the conditions that promote the spontaneous switching of thought towards matters unrelated to the task in hand, the imposition of a cognitive load decreases the frequency of task unrelated thought (e.g. Antrobus, 1968; Giambra, 1995). It is clear that theoretical explanation of these findings at the level of resources alone is unsatisfactory as their is no apriori reason that the imposition of cognitive resources will lead to either and increase or a decrease in task unrelated thinking.
A more useful level of analysis should build on similarities in this field of  research.  One such area in which parallel conclusions can be drawn is the processing of self-relevant material (Klinger et al. 1978; Petrie et al. 1998). Accounts of the self often focus on the role of a distributed resource (Tesser, 2002; Teasdale and Barnard, 1993) a perspective consistent with Dennets (1987, 1991) account of intentionality.  Empirical evidence which has examined the contrast between a resource based account and a distributed account of the frequency of TUT brought the relationship between TUT and resource allocation in line with the results of the thought suppression paradigm (Smallwood et al in press A & B; Smallwood et al. in submission).
In conclusion a more complex notion of resource allocation is required when researching thinking and this notion might reflect the interaction between a distributed account of thinking, reflecting the accumulation of our experience (Miller 2000) and an effort driven resource.  In such a framework effort might be better considered as an affordance (Gibson, 1979) rather than an objective criterion.
 
 

Carlos Souza
Acquisition of linguistic competence in infants (7 to 13 months)

In general the studies on language acquisition have been focusing in the structural aspects and from a perspective that language is a unitary phenomenon. Some few ones recognize its multiple character and they have been trying to analyze the functional aspects of this process. A proposal of functional analysis of language acquisition, based on Wittgenstein's language games notion, have been used to verify as the children learn different linguistic competences. At first some studies indicated that 2 year-old children learned the competences of choosing under linguistic criterion and of denominating, only due a systematic exposition to word-object pairing (i.e. without the reinforcement of child's responses).  The pairing's effects seemed to be modulated by the pairing's number, the order of presentation of the word-object pairings and by the interval between the pairing and the "acquisition tests". Besides, the children of this age seemed already to have acquired the competence of repeating vocalizations as a generalized class of responses. These results suggested the need of accomplish longitudinal studies with infants (age under 12 months).  Thus, two longitudinal studies tried to investigate as children 1) through the 7 to 10 months learned the competence of choosing under linguistic criterion; and 2) through the 10 to 13 months learned the competence of denominating. In an interaction situation with the teachers, the children were exposed to 5 word-object pairing per day, three times a week, through the three months of the study. It was verified that by means of the word-object pairing the children didn't learn the competences. However, it was observed in the two studies the gradual acquisition of gazing's responses and, gesture and vocal imitation. These results are discussed in terms of the function of the word-object pairing, of the reinforcement of the responses and, of the precurrents behaviors in the process of acquisition of the linguistic competences.
 

Manuel Sprung
University of Salzburg (Austria)

Children’s Understanding of Point of View: Belief and Sortals.

Doherty and Perner (1998) reported a very strong and robust correlation between children's ability to understand false belief and their understanding of the use of synonyms in a "say something different" task. In the belief task children have to understand that a person who doesn't witness an unexpected transfer of an object will mistakenly search in its original location. In the say-something-different task children played a game with familiar synonymous expressions, in which they named an item (e.g., "TV") and a puppet had to name it correctly with a different name (e.g., "television"). On one trial the puppet did it correctly, on two other trials the puppet made mistakes, either using the same word as the child ("TV") or use an inappropriate label (e.g., "table"). Children passed the task if they judged the appropriateness of puppet's response correctly in each case. As many children passed this task as did the FB task and the two tasks correlated strongly (even with age and verbal intelligence partialled out). The authors' explanation focused on the fact that both tasks require understanding of representation, i.e., that the believer misrepresents the location of the object, and that different words can have the same representational content (meaning).
 The present study uses the same methodology but extends the work by using basic versus super-ordinate category pairs (e.g., if child calls it "a bird", then puppet has to call it "an animal") and three other variations: name-colour (bird – yellow), colour-colour (one colour – other colour of two coloured object), and part-part (head – tail of dog). Results from 37 children aged 3;5 to 4;10 show the same results for basic and super-ordinate category pairs as for synonyms in the original work.  About as many children passed the false belief task (FB) as the basic vs. super-ordinate category naming task (NN) and these tasks were strongly correlated even after age and verbal intelligence (K-ABC) were partialled out (see Table 1). This finding makes it difficult to maintain the explanation given by Doherty and Perner because, unlike synonyms, basic and super-ordinate categories do not have the same meaning.
 The result is also difficult to explain in terms of a similarity of task complexity because performance on the other three say-something-different tasks was much superior (Table 1; NN vs. NC: McNemar’s ?2(N=18) = 8.0, p< .01) even though they pose the same problems of memory and rule understanding as the NN task. This difference becomes especially clear when looking only at the 15 children below 4 years: Only 20% mastered FB and 27% the NN task, whereas over 80% mastered each of the three other tasks. The problem with the NN task (and the synonyms task used earlier) is reminiscent of the "mutual exclusivity" principle in word learning (Markman, 1989). Children are reluctant to acknowledge that one object can be two things at the same time. However, it remains unclear why this initially useful but later impedimental principle should be overcome with the understanding of false belief. We discuss the possibility that different names (sortals) for an object individuate the object differently (Hirsch, 1982) yielding different perspectives (such that different propositions hold true). This could explain why this task is mastered at the same time as the FB task as it also requires the understanding and coordination of perspectives.

References.
Doherty, M. J., & Perner, J. (1998). Metalinguistic awareness and theory of mind: Just two words for the same thing? Cognitive Development, 13 , 279-305.
Hirsch, E. (1982). The concept of identity. New York, Oxford: Oxford University Press.
Markman, E. M. (1989). Categorization and naming in children: problems of induction. Cambridge, MA: A Bradford Book.
 
 

Julia Tanney
University of Kent (UK)
Conceptual Analysis, Theory Construction, and Conceptual Elucidation

Wittgenstein’s views have largely been ignored in contemporary work in the philosophy of mind because of the widespread view that he was simply prejudiced against a scientific understanding of the mental.  But some of his ideas are indirectly creeping into contemporary discussions.  This paper discusses one point of contact.  It notes the close relationship between traditional conceptual analysis and theory construction with regard to the attempts in recent philosophy of mind to construct a theory of content.  The paper argues that the same problems that plague the “classical” conception of (mental) concepts also infect the idea that a body of knowledge underlies our ability successfully to deploy mental concepts in commonsense psychological explanations.  If conceptual analysis and theory construction go hand in hand, then perhaps philosophy conceived as conceptual elucidation in the manner practiced by Wittgenstein will be turn out to be the most fruitful way of investigating the mental.
 
 

Dario Taraborelli
Institut Jean-Nicod, Paris (France)
Feature Binding and Object Perception. Does object awareness require feature conjunction?

Recent work in different fields of cognitive sciences seems to support the idea that in order to explain awareness of visually presented objects a subject needs some kind of binding mechanism, responsible for the conjunction of  different sensory features into a whole. Selective attention is commonly invoked as the key for solving this conjunction problem. Accordingly, the cognitive neurosciences have begun to investigate what kind of neural processing could underlie such a process. We analyse the evidence provided for justifying the existence of a binding problem and we argue against the claim that there is a feature binding problem to be solved in order to explain unity of object awareness. In particular, we question the definition of ‘feature’ and suggest some possible sources of misunderstanding related to this definition. Finally, we suggest an alternative approach in which perceived object unity does not rely exclusively on attentional conjunction of sensory features: instead of a general and unique mechanism mediating object awareness, we examine evidence that there are at least as many binding mechanisms as potential ways of interacting with visually presented objects.
 

Mark Textor
King’s College, London (UK)
How One Should Become A Sensible Millian Heir

Contemporary Millians argue for the thesis that the meaning of a proper name is its bearer by relying on intuitions about propositional attitude ascriptions. This argumentative strategy is found wanting and an alternative is proposed: An answer to the question «What is the utility of a proper name?» will give us a less problematic general premise for Millianism and it will point to a more sensible version of this doctrine that can respect Fregean intuitions.
 
 

Anne Tüscher
Institut Jean-Nicod, Paris (France)
The seeing-in theory of depiction and the psychophysics of picture perception

In this paper, I will examine Richard Wollheim's seeing-in theory of picture representation, and particularly one of the most convincing arguments in its favour, in the light of the recent cognitive psychology. Whereas Wollheim's theory is based mainly on conceptual and phenomenological considerations, I would like to review the psychophysical facts more closely in order to find out whether they have any implications for his argument.
I begin with an account of Wollheim's theory and of its major conceptual consequence: the claim that picture representation necessarily involves a simultaneous and conscious twofold experience of the picture's design and its content. Then I explain Wollheim's major justification for "twofoldness": the existence of a perceptual constancy phenomenon in picture perception, i.e., the fact that observers do not seem to be bothered by distortions when viewing a picture from another vantage point than the centre of projection. Afterwards, I move to the main and most discussed psychophysical explanation for the perceptual constancy phenomenon: the compensation hypothesis. On such an explanation, we are somehow correcting the optical distortions by means of a perceptive mechanism enabling us to reconstruct the pictured spatial layout as if the latter was seen from the correct viewpoint. This mechanism seems to be dependent on an awareness of presence of the picture's surface. Lately, this hypothesis has been challenged by new experiments. I will expose the major amendments leading to a new theory: the modified theory of compensation in picture perception, first put forward by Yang & Kubovy in 1999. Finally, I examine various consequences of this new version for Wollheim's twofoldness thesis and conclude that the latter survives the challenge of the new findings.
 
 

Agustín Vicente & Fernando Martínez-Manrique
Universidad del Pais Vasco (Spain)

Linguistic Relativity and the Language of Thought

Linguistic relativity is the idea that differences between languages generate corresponding differences between kinds of thoughts in people who speak each language. Our purpose is to reevaluate some relativistic theses in the context of Fodor’s defense of the existence of a language of thought. Fodor has provided the most important and best-known arguments against the cognitive use of public language. The idea that thought has its own language, Mentalese, universal and independent from public language, has become the orthodox view. However, as his views have developed, adding a general account of the structure of mind (the modularity thesis) and a psychosemantic theory (informational atomism), matters have become more complex. In this paper we will review some of his key arguments for the priority of thought over language.
We argue that Fodor’s position involves a variety of lexical relativity, i.e., the idea that the lexicon of public language is mirrored by the conceptual elements in thought. In his early work, Fodor acknowledges a mild form of this relativity, limited to the concepts that are actually used, not to those can be possessed. But thought still appears as prior to language. Yet, when we review the learning and systematicity arguments in the context of his subsequent work, we find weaker support for the priority of thought. Indeed, Fodor’s latest psychosemantical theory, informational atomism, fosters a stronger form of lexical relativity perfectly compatible with the idea that language can set limits to possible concepts, not only to actual ones. Finally, we examine his most recent attempt to support the explanatory priority of thought, based on the compositionality that thought, but not language, exhibits. We contend that this argument clashes with the systematicity argument and faces his theory with a dilemma. The solution, we suggest, comes from the distinction, that Fodor is reluctant to make, between a semantical and a conceptual level.
 
 

Tillman Vierkant
Max-Planck Institute for Psychological Research (Germany)
Is There A Self? Two Crucial Distinctions

The self has become one important focus of many cognitive science debates today. Unfortunately though, the concept of self with its long philosophical history and its use in everyday language is so confused that, despite its obvious importance, it is not clear whether using it genuinely solves any problems, rather than generating an additional terminological muddle. This paper tries to bring some light into the terminological darkness. It argues that there are two important distinctions in the usage of the term self. The first distinction is Metzinger’s distinction between subject and self-model. His subject is a naturalised version of the Kantian noumenal self, or in other words: an acting cognitive system. The self model, on the other hand, is a representation which the subject has of itself. The investigation agrees with parts of Metzinger’s analysis, but at the same time it tries to make clear that it is a mistake to reduce talk about the self to talk about the self-model. Such a tendency neglects the equally deeply rooted intuition that selves are subjects in Metzinger’s sense. If this distinction is used carefully, it can do some work in clearing up some terminological misunderstandings, but on its own this is not sufficient. The necessary second distinction is based upon Bernhard Williams’ idea of type and token persons. He argues that we are never sure in our evaluations, whether we think that a certain complex of abstract properties (type) or the actual material instantiation of these properties (token) is what matters for us in ascribing personhood. The paper uses this terminology to defend a similar distinction. It holds that in the current debates around the self there is a type and a token perspective on the self. The token perspective focuses on the instantiation of a concept in (self model) or as (subject) an organism, whereas the type perspective understands the self as a set of abstract properties in a narrative web. The major difference between Williams’ usage and the usage in the investigation is that the latter contends that the type perspective on selves nowadays has a distinct social constructionist background. Finally, it is demonstrated how the two distinctions help avoiding terminological problems in the psychological discourse around the narrative self.
 

Oscar Vilarroya
Universitat Autonoma de Barcelona, (Spain)
A Sensational Crisis

Psychologists interested in perception have traditionally distinguished between the notion of « sensation » and that of « perception ». This distinction has been usually based on intuitive criteria, rather than based on a fully developed theory. The central point of the distinction points to differentiating between some sort of neurophysiological state, which is the result of the senses being affected by some stimuli (sensation), and the result of a conceptual analysis of such a state (perception). In this paper, I present some research that is being conducted in developmental neurobiology, neurophysiology and cognitive neuropsychology which suggests that the processes involved in perception are so intermingled that there is little value in trying to divide them up neatly into sensation and perception, at least in the way that psychologists have proposed. First, I show that perception is influenced by a top-down flow of information, which intervenes in many different functional directions; in other words, there is a strong processing of feed back processing of information between final and initial stages of the perceptual process. Secondly, I show how crossmodal sensory integrations is not only a fact, but it may be even a requirement.  Finally, I present cases in which higher cognitive activities, such as categorization, are performed in the very same perceptual processing. I conclude by arguing that there could still be a role for the notion of « sensation ». I defend the idea that the traditional philosophical analysis about sensations is more fit to account for present neurobiological research than the psychological approach. A sensation would then correspond to what is traditionally labeled as sense-data: the objects that are experienced in a perceptual process; the contents of a perceptual experience. A perception would be the conceptual account of a certain sensory experience, and such a sensory experience would have a neurophysiological account.
 
 

Sven Walter
Ohio State University (USA)
Multiple Realizability and Property-Identities: An Incompatible Couple?

I will argue that contrary to Putnam, Fodor and the received wisdom in current philosophy of mind, the multiple realizability of mental properties does not rule out the possibility of the psychophysical property-identities required by the identity-theory.
The problem with the usual formulations of the argument from multiple realizability can be put in the form of a dilemma: Either the argument can be backed up by a fairly plausible, intuitively convincing story about multiple realizations of mental properties in different species, but then it only allows the derivation of a conclusion too weak to refute the identity-theorist’s claim of psychophysical property-identities. Or the argument is amended with an additional premise which allows the derivation of a conclusion strong enough to rule out the possibility of psychophysical property-identities, but then it loses most of the plausibility and intuitive force of the original argument and is no longer well-supported. Either way the argument from multiple realizability does not show that the multiple realizability of mental properties is necessarily incompatible with the psychophysical property identities required by the identity-theory.
 
 

Tiziana Zalla, Nelly Labruyère, Nicolas Georgieff
Institut des sciences cognitives, Lyon (France)

Goal-directed action representation in autism

Autism is a developmental disorder characterised by severe difficulties in communication, social interaction and executive functions. According to Leslie (1987; 1991) et U. Frith (1990) an impairment of the “Theory of Mind”, an innate functionally specialised system which allows the acquisition of mental concepts and the construction of metarepresentation, would be the core deficit of autism. A number of studies have shown that autistic individuals are impaired in understanding intentional stories despite a preserved ability to understand stories depicting behavioural or mechanical events (Baron-Cohen et al., 1985).
The aim of the present study is to investigate the ability to represent goal-directed actions in subjects with autism. Goal-oriented actions differ from both mechanical and intentional events in that they are represented in memory as sequences of structured events sharing the same goal. According to our hypothesis, the cognitive processes subserving goal-directed action representation would be impaired in autistic individuals.
In the present experiment, autistic subjects (n=16) were presented with series of pictures depicting component actions of three types of complex events: (a) short-term and (b) long-term goal-directed actions, and (c) mechanical stories. They were requested to reconstruct the correct event sequences under two conditions. In the mixed condition, children ability to sort the relevant events was challenged by presenting concurrently actions belonging to four different familiar activities. In the no-mixed condition, the same sequences were presented separately in order to minimise working memory demands. Their performance was compared with those of mentally retarded subjects (n=15) and normal subjects (n=16). Autistic children were impaired in organising goal-directed events (both short-term and long-term sequences) in their correct temporal order in both conditions (mixed and no-mixed), while they had no difficulty in ordering the mechanical sequences.
The present study indicates that an impairment in the ability to represent human actions as structured sequences of goal-oriented events might contribute to explain social and executive dysfunctions in autism.
 

Sponsors :

Conseil général du Rhône
Région Rhône-Alpes
Pôle Rhône-Alpes des sciences cognitives
Université Lyon 1

Many thanks to :

Institute of Cognitive Science
Ecole Normale Supérieure
Institut Jean-Nicod
Délégation régionale CNRS
Sylvie Jacquot

Institut des Sciences Cognitives (Institute of Cognitive Science)
UMR 5015 CNRS  UCB Lyon 1
67, boulevard Pinel 69675 BRON cedex
33 (0)4 37 91 12 12   33 (0)4 37 91 12 10 web@isc.cnrs.fr

ACCUEIL ISC
ICS English index