On this paper, Churchland explores folk psychology.

It is clear that any human can explain the behavior of other persons with remarkable accuracy. Each understands others because we share knowledge about the relations between external circumstances, internal states and behavior. This is called folk psychology.

The other minds problem states that the conviction that an individual is subject of certain mental states is not inferred deductively from his behavior. Rather, it is a explanatory hypothesis that provides explanations, predictions and understanding of the individual’s behavior. A Martian could do this, even if its own psychology was wildly different from ours, and he wouldn’t be generalizing from his own case.

The attitudes described in folk psychology can be compared to the ‘numerical attitudes’ shown by physical objects in terms of mass, velocity etc. which allow for generalization across attitudes that hold in nature. Therefore folk psychology is a theory, and it is striking that philosophers took so long to realize so. This means that the body-mind problem can be realized in terms of how the ontology of the folk psychology theory will be related to the ontology of neuroscience.  So will folk psychology be reducible in terms of another theory, perhaps a better one? Continue reading ‘Exploring Folk Psychology [notes on cognitive science]’

It is hard to think why random squiggles on this screen are triggering what we call “mental pictures” in your head. Putnam tried to account for this on his essay “brains in a vat”. More after the jump:

Continue reading ‘Brains in a Vat [notes on cognitive science]’

Dennet had the great idea of “formalising” folk psychology by ascribing beliefs, desires and intentions to objects in order to describe and predict their behavior. This is called the intentional stance, in which one predicts the most rational behavior of the object described according with the current beliefs held by it. This means assigning the object the beliefs it ought to have according to its perceptual abilities; the desires it ought to have given its biological needs and finally the behavior that would be rational given the beliefs and desires it holds.

This sounds very good, because we make no commitments about the internal workings of the system nor its design or program. Therefore, if we have two robots designed to act exactly like you, mimicking every single action you do, in the context you do them; they wouldn’t be any different from each other or from you from an intentional point of view. This means that describing an object as an intentional system does not entail anything about its physicochemical nature or functional design, and neither neurophysiology nor ”subpersonal cognitive psychology” could possibly show that the object is not an intentional system. These new concepts are identical with folk psychology, but they are better, since they capture its essential notions. These must be comparable to the older folk notions when it comes to describing ourselves.

However, accepting the intentional stance notions of Belief and Desires (which we will call ISBelief and ISDesire) would not allow us to describe human behavior, since a great deal of the time humans behave quite irrationally. However, the intentional stance assumes perfectly rational systems so attributing inferential failings and inconsistent beliefs to them makes no sense. How would you explain suicides, smokers and so forth? Dennet takes two routes. The first, called by Stich “The hard line” sticks to the idealized notion of intentional systems and tries to minimize the importance of the differences between IS beliefs and intentions and their folk-psychology counterparts. He uses arguments springing up from evolutionary optimality: if an organism is the product of natural selection, we can assume most of its beliefs are true and their belief generation strategies rational. The “soft line” legitimizes human faliability, by deriving “imperfect intentional systems” from the idealized ones explained above. However, this would be like adding a tolerance rule to chess such that any game can be considered legal as long as there are at most k illegal moves of chess.

Stich aims to step away from Dennet’s theory by reshaping it. There are two main points. First, rather than an instrumentalist view of folk psychology, he takes a functional one: belief states are functional states which play a role in the causation of behavior. Second, in ascribing content to belief states, we are not comparing others to an idealistic rational entity, but to ourselves. This makes it possible for three people reading about an earthquake on different sources have the same belief that the earthquake actually happened, and it accounts for the fallibility of human reasoning.

from “Dennet on intentional systems” by Stephen P. Stich

Phew. I’ve read all the slides a couple of times. I’ve written my own notes. I now see acronyms everywhere.

I’m revising for Multi Agent Semantic Web Systems. Since there are so many acronyms in this bloody course, and it is a bit hard to get one’s head around all of them, I’ve made a list. Lists are good.


  • MASWS: Multi Agent Semantic Web Systems.
  • MAS: Multi Agent System.
  • URI: Uniform Resource Identifier. Looks like a URL, but doesn’t have to be addressable (it’s nice if it is, especially if what you retrieve is a representation of the resource).
  • URL: Uniform Resource Locator.
  • RDF: Resource Description Framework.
  • XML: eXtensive Markup Language.
  • DC: Dublin Core. Resource description convention originating from the world of curation and libraries.
  • N3: A way of expressing RDF statements as triples followed by a period.
  • DTD: Document Type Declaration.
  • IR: Information Resource. Any resource that can be captured in a message (like a weather report).
  • SPARQL: (recursive) SPARQL Protocol and RDF Query Language. There are similarities between RDF and conventional relational databases. SPARQL is to RDF what SQL is to them.
  • JSON: JavaScript Object Notation.
  • SOAP: Simple Object Access Protocol.
  • WS: Web Service. A software system designed to support  Machine-To-Machine interaction.
  • WSDL: Web Service Definition Language.
  • RPC: Remote Procedure Call.
  • UDDI: Universal Description, Discovery and Integration. Repositories where web services are listed in a yellow-page fashion.
  • KQML: Knowledge Query and Manipulation Language. “outer” language that defines the format of messages exchanged between agents.
  • KIF: Knowledge Interchange Format. Lisp-like notation of First Order Logic, used to define the content of messages exchanged between agents.
  • FIPA: Foundation for Intelligent Physical Agents. Bringing a robot-war scenario to your world, many agents at a time.
  • FIPA-ACL: ditto – Agent Communication Language . Robots will speak this amongst each other while they argue upon the best way to erase the human race. It is nice because it allows for representing commitment. To kill.
  • FSM: Finite State Machine.

I’m not done, but i’m going to bed.


Sweet recipe passed down to me by Claudia, Michal’s girlfriend.

Continue reading ‘Polish Peach Cheesecake’

Necesitamos un nuevo juego del Mario, en el que rescatas a la princesa en los primeros diez minutos, y te pasas el resto del juego intentando reprimir esa sensación en tu estómago que te sugiere que ella esta trastornada, un detalle repetido una y otra vez en el curiosamente antisexual libro de instrucciones, y cuando Luigi bromea sobre ella y el Bowser, le rompes la nariz e inmediatamente te arrepientes. Cuando la princesa te pregunta, en el silencio del dormitorio del castillo del champiñon “¿Aún me quieres? tu te haces el dormido. Apretas el boton A ritmicamente, intentando controlar tu respiracion, manteniendola uniforme.



frontbackUna selección en español de

indie iberico

Gracias a angelocatil por las fotos.