Folk Psychology is Here to Stay [notes on cognitive science]

05May09

Terrence Horgan and James Woodward defend Folk Psychology on their paper “Folk Psychology is Here to Stay” from the attacks of Churchland and Stich, arguing that none of them has provided convincing reasons for doubting Folk Psychology as a method of explaining individuals. They base their critique on the overly stringent conceptions of the interactions between FP and other lower level theories. They define FP as consisting of two components: theoretical principles (universal closures of conditional formulas) and an existential thesis (that our FP descriptions of people are true and they do undergo the FP events attributed to them). Horgan and Woodward claim that Stich and Churchland argue against the existential thesis, by denying the existence of beliefs, desires and intentions. Horgan and Woodward thus claim that FP provides a useful framework for prediction and genuine causal explanations.

First, they recap the three considerations of Churchland in regard to the incompatibility between FP and neuroscience:

  1. FP suffers explanatory failures on an epic scale
  2. FP has been stagnant since the time of the Greeks
  3. FP’s categories stand alone, without a prospect of reduction to neuroscience.

Churchland’s irreducibility point is supported by the previous two.

Let us explore Horgan and Woodward’s critique of Churchland’s three points. First, they counter the lack of explanatory power of FP in the realm of learning, illnesses, etc with the fact that cognitive psychologists have developed successful theories of vision, memory and learning which use concepts directly extracted from Folk Psychology, such as desire, belief, judgement, etc.

Second, they criticize Churchland’s conception that there is an a priori set of phenomena that any successful psychological theory must account for in an unified way.  The fact is that knowing which set of phenomena a theory must account for requires considerable theoretical knowledge; something we lack in the case of psychological phenomena. There is no good a priori reason that should demand a theory designed to explain common human actions in terms of beliefs, desires and intentions to also describe visual perception or muscular coordination.

Horgan and Woodward go on to explore the stagnation of folk psychology. They argue that FP has actually changed significantly and progressively. They give the example of 18th and 19th century Europeans explaining human behavior in terms of character types, as opposed to 20th century europeans, who use situational factors. They compare the falsification of the FP theory in terms of stagnation to rejecting a physical theory of forces in terms of one’s notions of pushing and pulling remaining empirically unprogressive. They also point out that the user of this theory will be more interested in making particular causal judgements than formulating new causal generalizations.

Third, they criticize Church’s irreducibility claim, acknowledging that while reduction would be one way of salvaging FP, this reduction is an unlikely prospect given the fact that FP is 25 centuries old. And even if this reduction is impossible, and other theories provide good low-level account of homo sapiens behavior, it doesn’t mean that FP has to be false. They go on to mention Davidson’s alternative view called anomalous monism where the token mental events of FP can be linked to each other according to FP. Churchland avoids mentioning this theory.

On the second part of their article, they consider Stich’s claims of FP’s incompatibility with a developed CS. Stich offers two arguments.

The causal organization of the cognitive system probably does not conform with the causal organization which FP ascribes to it. Stich uses experimental evidence to claim that psychological events which control non-verbal behavior are independent of those which control verbal behavior. If evidence in CS generalizes the view that there are independent cognitive subsystems for the control of verbal and non-verbal behavior then we will be forced to conclude that there are no such things as beliefs. He uses Storms-Nisbets experiments on insomniac subjects given placebo pills with opposite effects, concluding that the mechanisms that account for non-verbal and verbal behavior are largely distinct. He goes on to produce three conclussions:

  1. In the cases described there is no consistent way to ascribe beliefs and desires since the cognitive causes of the non-verbal behavior are distinct to the causes of verbal behavior.
  2. It is likely that in general the systems that control verbal and non-verbal behavior are distinct.
  3. It is likely that FP is radically false.

Horgan and Woodward dispute these in turn. First, they appeal to unconscious beliefs to explain the lack of verbal expression of the beliefs in the studies explored. They explain that Storms and Nisbett’s studies establish a hierarchy of beliefs, where unconscious beliefs might or might not elicit conscious beliefs, and those (which constitute hypotheses about the causes of the original behavior) cause the subsequent verbal behavior.

Second, they negate 2 by explaining that while unconscious beliefs may elicit non-verbal behavior in a large number of cases, it does not prove that this is the case in all situations. If this was the case, engaging in a complicated task while explaining what you are doing would be impossible. Therefore, the “dual-control” argument against FP is not successful.

On the third part of the article, they discuss the modularity principle of FP, that is, that the building blocks of FP should be identical with naturally isolable parts of the cognitive system. Stich points out that on the recent yeas theorists have become skeptical about highly modular memory models, due to the amount of non-deductive inference necessary for language use and comprehension. But if this is the case, then not only will FP fail to map with lower-level theories, but CS with neuroscience, neuroscience with physics-chemistry and so on. So, the modularity priciple, as Horgan and Woodward point out, is largely implausible as an inter-theoretical compatibility condition.

Neither Stich nor Churchland justify the modularity principle nor the demand for reducibility respectively. One reason why Stich might have imposed the modularity principle is the doubt of whether propositional attitudes involve an internal representation at all. Horgan and Woodward try to find support for the modularity principle in the Davidsonian considerations of the role of laws in causality. These state that

  1. FP contains only heteronomic generalizations rather than strict laws
  2. Two events are related as cause and effect only if they have descriptions which instantiate a strict law
  3. Event descriptions which instate a strict law must be naturally isolable from the perspective of that theory

However, Horgan and Woodward prove that the Davidson argument for modularity principle has a false premise, so we can reject the need for modularity altogether.

Part four of the paper deals with the possible parallelism between FP-events and CS-events. For the non modular vision of FP and CS to be plausible then it is essential that events can be considered causes. They define minimal causation as events that while they might be a genuine cause of another, there is a smaller part of the event that causes the same event. If an event causes two events F and G, the parts of the original cause that give rise to F and G might be different. They conclude that FP could be true even if modularity is violated, where FP-events can cause their effects in a conglomerative manner, with different effects having different parts of the event as their minimal causes.

Finally, part five of the paper considers the objection that leaving out the modularity principle would result in too lenient an analysis, allowing upper-level causal claims to come out true even if they seem inconsistent with the underlying theory. Falsity of an upper-level theory can be detected by its failure to fit with an underlying theory with we have a strong reason to believe to be true. Horgan and Woodward’s explore the case of possession by the devil, which is to be understood in terms of the role it plays in a network of causal relations. The pathological behavior arising from possession by the devil will likely be eliminated by the use of exorcism, but not by the use of drugs or psychotherapy.  Now if this event d of being possessed by the devil is the cause of some behavior, there must be an event e describable by an underlying theory such that d is identified within it. It is easy to see how finding this sort of event is impossible from our current physical and chemical theory.

In sum, Horgan and Woodward conclude that their non-modular approach is appropriately permissive, as opposed to the overly strict modular view. Thus the failure of behaviorists at uncovering stimulus-response laws that undercut the causal architecture of FP, the fact that FP lies at the core of cognitivist theories and that FP serves us well to explain and predict behavior means that “folk psychology is here to stay”

from “Folk Psychology is Here to Stay” by T. Horgan and J. Woodward

Anuncios


No Responses Yet to “Folk Psychology is Here to Stay [notes on cognitive science]”

  1. Dejar un comentario

Responder

Introduce tus datos o haz clic en un icono para iniciar sesión:

Logo de WordPress.com

Estás comentando usando tu cuenta de WordPress.com. Cerrar sesión / Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión / Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión / Cambiar )

Google+ photo

Estás comentando usando tu cuenta de Google+. Cerrar sesión / Cambiar )

Conectando a %s


A %d blogueros les gusta esto: