Krivo Flores A Response to Fodor 20 I find the above statements rather extraordinary in that they seem to be points at which Fodor is quite obviously mistaken as well as where his admitted dearth of Hume scholarship becomes evident Fodor, p. Fodor is mistaken both in his assertion that Hume had no notion of unconscious ideas and his assumption that the imagination lacks access to a structured memory, as clearly evinced by the following passages which at once appear to contradict both claims by Fodor; many similar examples could be offered: It will here be worth our observation, that the past experience, on which all our judgments concerning cause and effect depend, may operate on our mind in such an insensible manner as never to be taken notice of, and may even in some measure be unknown to us.
A person, who stops short in his journey upon meeting a river in his way, foresees the consequences of his proceeding forward; and his knowledge of these consequences is conveyed to him by past experience, which informs him of such certain conjunctions of causes and effects [ The objects seem so inseparable, that we interpose not a moment's delay in passing from the one to the other. But as this transition proceeds from experience, and not from any primary connexion betwixt the ideas, we must necessarily acknowledge, that experience may produce a belief and a judgment of causes and effects by a secret operation, and without being once thought of [ Fodor might respond that the above quotes merely show that Hume allowed unconscious beliefs, judgements, or dispositional habits, but not unconscious ideas.
Judgement, which the above quote also asserts may occur unconsciously, is itself according to Hume, the forming or uniting ideas even a single idea therefore judgements too may comprise unconscious ideas. For if it were true as Hume and Berkeley claimed, that the contents and structure of abstract ideas derives externally, from public language i. To which Hume may readily retort: Of these impressions or ideas of the memory we form a kind of system, comprehending whatever we remember to have been present, either to our internal perception or senses; and every particular of that system, joined to the present impressions, we are pleased to call a reality.
But the mind stops not here. For finding, that with this system of perceptions, there is another connected by custom, or if you will, by the relation of cause or effect, it proceeds to the consideration of their ideas; and as it feels that it is in a manner necessarily determined to view these particular ideas, and that the custom or relation, by which it is determined, admits not of the least change, it forms them into a new system, which it likewise dignifies with the title of realities.
The first of these systems is the object of the memory and senses; the second of the judgment THN 1. Krivo Flores A Response to Fodor 22 Experience is a principle, which instructs me in the several conjunctions of objects for the past. Habit is another principle, which determines me to expect the same for the future; and both of them conspiring to operate upon the imagination, make me form certain ideas in a more intense and lively manner THN 1. Krivo Flores 3 Prediction Machines: The Imagination as Probabilistic Generative Model As we saw in the previous chapter Fodor claimed that Hume was trying to get the imagination to do a job for which he should have invoked traces, and once traces are admitted, says Fodor, a faculty of the imagination becomes otiose , p.
I now endeavor to demonstrate that this is precisely what the action-oriented predictive processing model promises Clark, ; Hohwy, Fodor claims that imagination and association, construed as cognitive faculties, are unnecessary because Turing showed that both can be replaced by standard computational means Fodor, p. Recall, the mind, according to Fodor is literally at the functional level a Turing Machine type of computer, as evinced by the following passage: [A]ssuming that the mind has the sort of architecture that Turing machines do [ A much more apt characterization of the human mind, however, and one which, I argue, also happens to accord greatly with Hume, is more along the lines of a Helmholtz machine rather than a Turing-machine.
For not only does the former conception accommodate a faculty of the imagination it requires one—in the form of a probabilistic generative model.
- Sams Teach Yourself Networking in 24 Hours (Sams Teach Yourself...in 24 Hours).
- Practical Immunodermatology;
- Topics in Mathematical Analysis and Applications.
- HUME: large-scale detection of causal genetic factors of adverse drug reactions.;
- Digital Systems Principles and Applications.
- Navigation menu.
- Asymptotic stability for nonlinear Kirchhoff systems;
Krivo Flores A Response to Fodor 24 3. It was a formative childhood experience which first led Helmholtz to a radical conjecture: that perception itself requires the brain to engage in continuous acts of unconscious inference. Helmholtz described the experience which led to his pivotal insight as follows: I can recall when I was a boy going past the garrison chapel in Potsdam, where some people were standing in the belfry.
I mistook them for dolls and asked my mother to reach up and get them for me, which I thought she could do. The circumstances were impressed on my memory, because it was by this mistake that I learned to understand the law of foreshortening in perspective Brockman, p. This early experience convinced Helmholtz that the passive information available from the retina and other sensory organs is itself ambiguous and insufficient for meaningfully bringing forth the world in perception Brockman, p.
Size, distance, orientation, and causal relations all have to be inferred from the sampling of ambiguous, uncertain and often conflicting sensory cues. Hume, as the following passage shows, like Helmholtz was also aware of the importance of such inferential judgements to perception: The judgement here corrects the inequalities of our internal emotions and perceptions; in like manner, as it preserves us from error, in the several variations of images, presented to our external senses.
The same object, at a double distance, really throws on the eye a picture of but half the bulk; yet we imagine that it appears of the same size in both situations; because we know that on our approach to it, its image would expand on the eye, and that the difference consists not in the object itself, but in our position with regard to it. And, indeed, without such a correction of appearances, both in internal and external sentiment, men could never think or talk steadily on any subject; while their fluctuating situations produce a continual variation on objects, and throw them into such different and contrary lights and positions EPM p.
Krivo Flores p. The key innovation of the Helmholtz machine was its employment of a generative model. Clark, a key advocate of the predictive processing model, summarizes the generative model concept as follows: A generative model [ A good generative model for vision would thus seek to capture the ways in which observed lower-level visual responses are generated by an interacting web of causes—for example, the various aspects of a visually presented scene. In practice, this means that top- down connections within a multilevel hierarchical and bidirectional system come to encode a probabilistic model of the activities of units and groups of units within lower levels, thus tracking [ Cognitive systems such as ourselves are constantly engaged in the precarious task of inferring sensory and worldly causes from their bodily effects.
Problematically, however, any such effect will be consistent with an indefinite number of possible causes and explanations. What is then needed is an effective strategy for biasing our system of prior experiential knowledge in a manner as to engender ever more accurate predictions. A task which, incidentally, requires the computing of multiple probability distributions, based on their relative, context dependent—and often subjective probabilities Clark, p.
These models suggest that representing the world in perception, critically depends upon canceling out sensory prediction errors Clark, p. Krivo Flores A Response to Fodor 26 temporal scales with top-down predictions which reflect what the system already knows about the body and world Clark, p. What an agent perceives thus depends heavily upon the set of prior beliefs its brain deploys in its best attempts to predict current sensory perturbations Clark, p.
Hume too showed an awareness of the role of prior knowledge and error correcting inferences in sense perception, as well as in our sentiments and language use as well: Experience soon teaches us this method of correcting our sentiments, or at least, of correcting our language, where the sentiments are more stubborn and inalterable [ The more perfect the match, the fewer error signals propagate up the neural hierarchy Clark, It is this process of prediction-error minimization operating hierarchically at multiple levels which is thought to enable the embodied brain to acquire and exploit a rich body of probabilistically couched knowledge at various levels of abstraction, to enable it to discern and represent the causal structure of the source domain of the signals perturbing it Clark, p.
Krivo Flores audio compression Clark, p. Clark explains the central idea behind predictive coding as follows: In most [digital] images, the value of one pixel regularly predicts the value of its nearest neighbors, with differences marking important features such as the boundaries between objects. What needs to be transmitted is therefore just the difference a. Though Hume was writing long before the era of JPEGs and data compression, he too emphasized the importance of expectation and surprise in cognition and perception, as evinced by the following: What is natural and essential to any thing is, in a manner, expected; and what is expected makes less impression, and appears of less moment, than what is unusual and extraordinary.
A considerable change of the former kind seems really less to the imagination, than the most trivial alteration of the latter; and by breaking less the continuity of the thought, has less influence in destroying the identity THN 1. And: When I throw a piece of dry wood into a fire, my mind is immediately carried to conceive, that it augments, not extinguishes the flame EHU 5. Predictive processing merely extends our notion of surprise, beyond the intuitive form, noted by Hume which occurs in conscious experience, to higher levels of abstraction and to lower neural levels of processing.
There is, however, a dark side to this unification of imagination, perception, and belief- formation. Krivo Flores A Response to Fodor 28 perception itself, according to this view, is determined largely by prior expectations, such may begin a cycle of misinformation allowing false perceptions and bizarre beliefs to become entrenched into an internally coherent closed cycle Clark, p.
While the pattern-completing nature of the predictive mind has the benefits of allowing perceptual systems to deal with noisy and incomplete information, it also—if the balance between top-down expectations and the driving sensory signal is improperly weighted or biased—opens the gateway to mental illnesses such as schizophrenia Corlett et al. As described in the above model, perceptions take shape as our brains endeavor to guess their own sensory e. Thus, a delicate balance is required to walk the fine line between perception, expectation, imagination, and hallucination, a delicate balance of which Hume was very aware: Nor will it be amiss to remark, that as a lively imagination very often degenerates into madness or folly, and bears it a great resemblance in its operations; so they influence the judgment after the same manner, and produce belief from the very same principles.
When the imagination, from any extraordinary ferment of the blood and spirits, acquires such a vivacity as disorders all its powers and faculties, there is no means of distinguishing betwixt truth and falsehood; but every loose fiction or idea, having the same influence as the impressions of the memory, or the conclusions of the judgment, is received on the same footing, and operates with equal force on the passions.
A present impression and a customary transition are now no longer necessary to enliven our ideas. Every chimera of the brain is as vivid and intense as any of those inferences, which we formerly dignified with the name of conclusions concerning matters of fact, and sometimes as the present impressions of the senses THN 1. Krivo Flores The merging of perception, causal inference, belief formation, and imagination is usually adaptive, but when things go awry, false inferences may spiral and feed back upon themselves leading us to merely see and what we predict and to believe what we expect.
At this point it will well serve my current endeavor to examine how the interpretation of probability as degrees of partial belief provides a fruitful connection between Hume and contemporary models of Bayesian causal inference. Krivo Flores A Response to Fodor 30 4 Hume in the Light of Bayes Much of the research carried out within the predictive processing paradigm is highly mathematical, especially central to such research is the mathematics of Bayesian probability and inference.
Nothing is more curiously enquired after by the mind than the causes of every phenomenon THN 1. All reasonings concerning the probability of causes are founded on the habit or determination to transfer the past to the future and to assume that like causes beget like effects 1. The Predictive processing model like that of Hume e. Krivo Flores minimize the errors in its predictions of expected sensory inputs Clark, p. In this way it is able to bring past experiential knowledge to bear upon its constant efforts to understand its incoming barrage of sensory signals and the world, and to construct a rich model of its physical and social milieu Clark, Exploiting such probabilistically couched knowledge then enables the generative model to make better guesses in the future Clark, p.
Before I continue, it should be superfluous to note that I am not somehow suggesting Hume surreptitiously conceived of these ideas in precisely the same manner as the modern paradigm I wish to advocate. It is my contention that Bayesian probability may similarly elucidate Hume. Krivo Flores A Response to Fodor 32 4. An example, I shall argue, when expressed in modern terms, suggests that the primary function of the imagination is to synthesize conditional probability distributions in virtue of its capacity as generative model. Consider the following passage by Hume from his chapter Of the Probability of Causes in the light of what has been learned thus far of predictive processing: When we transfer contrary experiments to the future, we can only repeat these contrary experiments with their particular proportions; which could not produce assurance in any single event, upon which we reason, unless the fancy melted together all those images that concur, and extracted from them one single idea or image, which is intense and lively in proportion to the number of experiments from which it is derived, and their superiority above their antagonists.
Our past experience presents no determinate object; and as our belief, however faint, fixes itself on a determinate object, it is evident that the belief arises not merely from the transference of past to future, but from some operation of the fancy conjoined with it.
This may lead us to conceive the manner, in which that faculty enters into all our reasonings THN, 1. I contend this is merely another way of saying that the imagination computes via simulation, a posterior probability distribution, more on this concept shortly. As we will see, however, the predictive processing model is able to specify the underlying mechanisms in far greater detail than Hume, but in a manner complimentary to his original formulation.
Krivo Flores 4. I say, trivially, because Hume most certainly did have a well developed notion of degrees of belief which is precisely how conditional probability density functions are interpreted in contemporary Bayesian inferential statistics Pearl, p. First, his interpretation of probability as degrees of belief and focus on inductive causal inference.
Second, Hume also had a strong conception of the ineliminable role of the simulations of the imagination in probabilistic reasoning and causal inference, hence the connection to generative models as well THN 2. Despite lacking modern formalisms however, I contend that Hume displayed remarkable foresight concerning all of the central tenets of rational belief revision which would later independently become Bayesian confirmation theory.
Consider the following famous passage: A wise man, therefore, proportions his belief to the evidence. In such conclusions as are founded on an infallible experience, he expects the event with the last degree of assurance, and regards his past experience as a full proof of the future existence of that event. In other cases, he proceeds with more caution: He weighs the opposite experiments: He considers which side is supported by the greater number of experiments: to that side he inclines, with doubt and hesitation; and when at last he fixes his judgement, the evidence exceeds not what we properly call probability.
All probability, then, supposes an opposition of experiments and observations, where the one side is found to overbalance the other, and to produce a degree of evidence, proportioned to the superiority. Krivo Flores A Response to Fodor 34 at no point am I suggesting that there was any direct influence of the work of the reverend Bayes on Hume, nor do I believe that Hume had any knowledge of the relevant formalisms, my claim is only that such follow naturally and logically from his informal and undeveloped insights.
Further, I will now show that there are good arguments in support of my position. As such, Bayesian methods provide an ideal means for reasoning with partial beliefs under conditions of uncertainty. Bayesian methods are widely employed in the predictive processing paradigm, and it is also widely assumed that the brain itself performs some form of approximate Bayesian inference Doya, Krivo Flores A Response to Fodor 36 therefore each part of the probability contributes to the production of the belief, each part of the possibility must have the same influence on the opposite side; the nature of these parts being entirely the same.
Hume Variations - Jerry A. Fodor - Google книги
The contrary belief, attending the possibility, implies a view of a certain object, as well as the probability does an opposite view. In this particular both these degrees of belief are alike. The only manner then, in which the superior number of similar component parts in the one can exert its influence, and prevail above the inferior in the other, is by producing a stronger and more lively view of its object.
Each part presents a particular view; and all these views uniting together produce one general view, which is fuller and more distinct by the greater number of causes or principles, from which it is derived. This synthesizing of many contrary events into an idea of the probability of an expected is in turn another way of describing a simulation. Moreover, the system Hume informally and implicitly of course developed can be shown to have been a powerful forerunner to a later system developed by Rudolf Carnap Mura, p.
Krivo Flores based only on a single observation, such as when sample consists of just one individual. Mura, cites the following more detailed passage as Hume's basic argument for the straight rule for singular predictive inferences: Though we give the preference to that which has been found most usual, and believe that this effect will exist, we must not overlook the other effects, but must assign to each of them a particular weight and authority, in proportion as we have found it to be more or less frequent. Here then it seems evident, that, when we transfer the past to the future, in order to determine the effect, which will result from any cause, we transfer all the different events, in the same proportion as they have appeared in the past, and conceive one to have existed a hundred times, for instance, another ten times, and another once.
As a great number of views do here concur in one event, they fortify and confirm it to the imagination, beget that sentiment which we call belief, and give its object the preference above the contrary event, which is not supported by an equal number of experiments, and recurs not so frequently to the thought in transferring the past to the future EHU Thus, according to Hume, the degree of a belief in a given event depends on the relative "vivacity" of the corresponding idea Mura, As a result, each possible outcome of a future prediction, ceteris paribus, is allocated a degree of belief proportional to the number of observed instances.
Mura contends that Hume sees the straight rule as the counterpart of the classic definition of probability of chances, viz. The relative chances are then synthesized in the imagination and projected into a single predicted future event based on proportions of past observed instances, an operation analogous to that of a predictive generative model Goodman et al,. I argue that Hume should thus be seen to have connected the cognitive and affective aspects of degrees of belief with the conative.
Beliefs for both Ramsey and Hume are action- oriented. Recall, according to Hume, all of our reasonings concerning the probability of causes are founded on our instinctual habit of projecting past experiences onto expectations of the future THN 1. Krivo Flores driven, by force of habit, to expect B given any future perception of A. To move away from the strictly logical interpretation of probability as in Carnap, to a subjective—yet quantifiable—interpretation thereof in terms of degrees of partial belief qua basis of action, Ramsey developed an ingenious interpretation which included reference to psychological notions like beliefs and desires.
When the chances or experiments on one side amount to ten thousand, and on the other to ten thousand and one, the judgment gives the preference to the latter, upon account of that superiority; though it is plainly impossible for the mind to run over every particular view, and distinguish the superior vivacity of the image arising from the superior number, where the difference is so inconsiderable.
We have a parallel instance in the affections. It is evident, according to the principles above-mentioned, that when an object produces any passion in us, which varies according to the different quantity of the object; I say, it is evident, that the passion, properly speaking, is not a simple emotion, but a compounded one, of a great number of weaker passions, derived from a view of each part of the object. For otherwise it were impossible the passion should increase by the increase of these parts.
Thus a man, who desires a thousand pound, has in reality a thousand or more desires which uniting together, seem to make only one passion; though the composition evidently betrays itself upon every alteration of the object, by the preference he gives to the larger number, if superior only by a unit. Krivo Flores A Response to Fodor 40 preferring the greater number depends not upon our passions, but upon custom and general rules. We have found in a multitude of instances, that the augmenting the numbers of any sum augments the passion, where the numbers are precise and the difference sensible.
The mind can perceive from its immediate feeling, that three guineas produce a greater passion than two; and this it transfers to larger numbers, because of the resemblance; and by a general rule assigns to a thousand guineas, a stronger passion than to nine hundred and ninety-nine THN 1. The Neuroscientist Karl Friston has extended the above ideas beyond rational cognitive actions in terms of beliefs, to include the actual bodily movements involved in carrying out any such activities.
A process summarized by Hawkins and Blakeslee below: Thinking of going to the next pattern in a sequence causes a cascading prediction of what you should experience next. As the cascading prediction unfolds, it generates the motor commands necessary to fulfill the prediction. Thinking, predicting, and doing are all part of the same unfolding of sequences moving down the cortical hierarchy. With the larger picture now in view, I return to our engagement with Fodor. A purpose which may be further elaborated in terms of Bayesian predictive processing: the imagination functions to update the beliefs it synthesizes in such a manner as to minimize the overall prediction error in its ceaseless efforts at causal inference.
Additionally, these same underlying generative mechanisms for prediction-error minimization underlie perception, action, and imagination, as well as cognition. According to Hume the chief function of imagination consists in combining the separate ideas of experienced probabilities into a synthetic novel simulation which, by a communication of vivacity proportionate to the relevant proportions, it raises up beyond the fictions of the imagination to be made the content of a belief.
A belief which in turn may constitute the basis for action. We may now ask: just what type of specification of the workings of the imagination would Fodor find satisfactory? What is the form such a specification should take , p. He gives us a clue towards the end of chapter 5. Krivo Flores need to fully specify the workings of the imagination by saying that, since such a specification is largely a matter of implementation, Hume could simply absolve himself by remaining agnostic on questions pitched at that level of description Fodor, p.
No doubt, the implementation of psychological laws is eventually by brain mechanisms. And maybe in the second, third, and fourth instances, too Fodor, p. To perceive in this way, notes Clark: [M]eans that perception, as it occurs in creatures like us, is co-emergent with at least a functional analogue to imagination. As specified by Fodor, these imaginative generative operations which enact perception, understanding, action, and imagination are implemented hierarchically i.
Krivo Flores A Response to Fodor 44 however, some of these operations may reach up all the way to consciousness in which case they constitute mental imagery, and are thus, at least somewhat available to introspection and phenomenological investigation Barsalou, p. Considering the preceding discussion, apprehending the interconnected nature of imagination and understanding will perhaps be the most intuitive. Generative models, by design, try to guess their own sensory inputs by attempting to auto-generate those very signals for themselves, which is another way to say by imagining them Clark, p.
The remainder of this paper will be devoted to an exploration of this suggestion and to making use of the insights thus uncovered to conclude our response and counter challenge to Fodor. Krivo Flores senses as well THN 1. I take such statements, as well as the following, to favor the idea that Hume would be congenial towards the idea of unifying the sensory perceptions impressions , causal inference, and prediction under the purview of the imagination, as he already had made speculations much to that effect: As to those impressions, which arise from the senses, their ultimate cause is, in my opinion, perfectly inexplicable by human reason, and it will always be impossible to decide with certainty, whether they arise immediately from the object, or are produced by the creative power of the mind [ It is also another way of describing the synthesis of token ideas.
Cognitive psychologist and computer scientist Noah Goodman has noted that the idea of a generative model naturally accords with the notion of mental simulation. Not only is a form of semantic empiricism compatible with the faculty of imagination, but it appears that the faculty implements understanding and thus supports a form of semantic empiricism. Krivo Flores A Response to Fodor 46 we can robustly match the incoming signal, moment to moment, with apt expectations predictions.
But to know that to know how the present sensory signal is likely to change and evolve over time just is to understand a lot about how the world is, and the kinds of entities and events that populate it. Creatures deploying this strategy, when they see the grass twitch in just that certain way, are already expecting to see the tasty prey emerge, and already expecting to feel the sensations of their own muscles tensing to pounce.
An animal, or machine, that has that kind of grip on its world is already deep into the business of understanding that world Clark, p. The ties between predictive processing, action, prediction-error minimization, and meaning, surprisingly, go even deeper, and do so in a way which, I argue, connects quite naturally with the paradigm of embodied cognition and embodied simulation theories of linguistic meaning and understanding Bergen, ; Barsalou, ; Flores, more of which I will have to say later on. For now it will suffice to note that these theories, are empiricist theories of meaning and understanding and that the action-oriented predictive processing model suggests that to make successful predictions about, and thereby to effectively interact with, the world, is in very large measure what is required to understand that world Clark, p.
One may next ask: what is the cognitive medium in which mental simulations are enacted? Hume allowed only two mental objects: impressions and ideas, themselves subspecies of perceptions. As noted in Chapter 1, amodal symbols are arbitrarily related to their referents and do not contain any specific sensory or descriptive content; perceptual symbols in contrast, are modal in that they retain specific information derived from the sensory modalities from which were derived, such as sight, smell, sound, and touch, etc.
Krivo Flores directly related to, and derived from, actual perceptions. The reactivation of perceptual symbols, e. As natural and traditional as it is to think of perceptual symbols in these ways, this is not the form they take here. Instead, they are records of the neural states that underlie perception. During perception, systems of neurons in sensory-motor regions of the brain capture information about perceived events in the environment and in the body. At this level of perceptual analysis, the information represented is relatively qualitative and functional [ Instead, related symbols become organized into a simulator that allows the cognitive system to construct specific simulations of an entity or event in its absence, analogous to the simulations that underlie mental imagery Barsalou, p.
I argue that Barsalou has effectively developed a modern implementation of a type of copy principle in which the same systems of neurons in sensory-motor regions of the brain capture information about perceived events in the environment and body which may be later re-activated in mental imagery, to simulate objects in their absence, and in thinking and reasoning more generally Barsalou, p.
Thus a visual idea will reuse inter alia the same parts of the visual cortex required in actual visual perception. Krivo Flores A Response to Fodor 48 The perception of something familiar in the environment, body, or introspection activates a simulation or situated conceptualization that contains it. Components of the simulation or situated conceptualization not yet experienced constitute predictions about events likely to occur, actions likely to be effective and introspections likely to result. Because simulations represent predictions, predictions are simulated in the same modalities in which they would be experienced.
Consequently, predictions can be readily matched to corresponding components [of a situation] should they actually occur Barsalou, p. And: When a situated conceptualization becomes active, it constitutes a rich source of prediction via this pattern completion inference mechanism. A situated conceptualization is essentially a pattern, namely, a complex configuration of multi-modal components that represent a familiar situation. When a component of this pattern matches something experienced in the environment, the larger pattern becomes active in memory.
The remaining pattern components—not yet experienced—constitute inferences, namely, educated guesses about what might occur next Barsalou, p. Projectivism is the propensity, noted by Hume, by which: [T]he mind has a great propensity to spread itself on external objects, and to conjoin with them any internal impressions, which they occasion, and which always make their appearance at the same time that these objects discover themselves to the senses.
Thus as certain sounds and smells are always found to attend certain visible objects, we naturally imagine a conjunction, even in place, betwixt the objects and qualities, though the qualities be of such a nature as to admit of no such conjunction, and really exist no where 1.
Dennett has exemplified this approach by showing how a Humean take on predictive processing naturally meshes with core concepts in the wider paradigm of embodied and enactive cognitive science, namely the concepts of affordances and Umwelt: Every organism, whether a bacterium or a member of Homo sapiens, has a set of things in the world that matter to it and which it therefore needs to discriminate and anticipate as best it can. Hierarchical Bayesian predictions accomplish this, generating affordances galore: We expect solid objects to have backs that will come into view as we walk around them, doors to open, stairs to afford climbing, cups to hold liquid, and so forth.
But among the things in our Umwelt that matter to our well-being are ourselves! We ought to have good Bayesian expectations about what we will do next, what we will think next, and what we will expect next! Dennett, p.
The idea of an affordance comes from J. For example: a book affords climbing to a mouse, but not to a human. Affordances are properties in the environment that matter to an organism, different organisms will therefore perceive quite different affordances in their environments.
Grass affords nourishment to cows but not to humans, cars afford driving to many humans—but not to ants. Action-oriented predictive processing—predictive projectivism—suggests that affordances are constructed via the fine- tuning of projected expectations of the inherent environmental possibilities perceived as available to each particular organism.
Dennett above has thus imported Gibsonian affordances, quite brilliantly into the predictive processing paradigm by way of an apt analogy to Humean projectivism. According to Gibson, in contrast, affordances are perceived directly, unmediated by any mental representations or inferential processes. Dennett appears to be advancing a novel alternative account of affordances: that this highly mediated and statistical mechanism I have called predictive projectivism just is the biological mechanism which creates the phenomenon of affordances.
This I think is the correct view, what needs to be surrendered is the Gibsonian commitment direct perception, at least in naive form. It is here that we find a much unexpected ally in Jerry Fodor. Krivo Flores 6. The Establishment Theory, also called the "information processing" view states that perception depends, in several respects upon inferences Fodor and Pylysyn, p.
Perhaps just as surprisingly, I agree—provisionally—with his critique of direct perception, despite an earlier publication which suggests the contrary see: Flores, It is his solution, which I feel is mistaken. However, the following discussion will sound a somewhat irenic note in the current debate. Fodor and Pylyshyn contend that perception is a representationally mediated process which is highly indirect which I accept, but would soften to somewhat indirect.
They assert that perception, qua inferential process, requires the following triumvirate of capacities: memory, representation, and computational mechanisms capable of transforming, according to fixed rules, representations which function as premises, into other representations—the perceptual conclusions. Gibson, as noted above proposed a hypothesis without these factors, hence direct perception.
According to Gibson, we directly perceive the invariant properties of the environment without any representational intermediaries. The question I and Fodor both have is: how does this work? Fodor points out: "affordances" presuppose direct perception, thus cannot be used to explain it. The story is that you infer the latter from the former on the basis of usually implicit knowledge of the correlations that connect them: Fodor and Pylyshyn, p.
Krivo Flores A Response to Fodor 52 mechanism which Fodor proposes as underlying the inferential processes required for perception come in the form of transducers. However, as Turing Machines are universal computers they may model any other type of machine, including Helmholtz machines, and or other probabilistic inference engines, indeed most connectionist and predictive generative models are implemented on digital computers—Turing machines!
Any algorithmically specifiable form of non- demonstrative inference e. Sensory inputs are thus converted into symbols of the language of thought and processed accordingly. As Barsalou has shown above cognitive processes do not need to detach modal information via transduction. Arthur Glenberg has combined these above ideas: affordances, perceptual symbols, and simulation into an empiricist theory of human linguistic understanding. Words are indexed to perceptual symbols.
Affordances are derived from the component expressions. The hypothesis states: ESH We understand language by simulating in our minds the types of experiences the language describes. We do so by using the same parts of our physiology which initially evolved to perceive and act in the real world Bergen, p. What this proposes, for example, is that we use our visual system not only to see actual objects, but also to interpret sentences about visible objects and, similarly, that we employ our motor system to construct motor images when we interpret verbs or sentences involving actions and movements.
The fundamental idea behind the embodied simulation theory is that when we hear or read language, our brain creates a simulation of the situation described using the same regions which would be required if we were to experience the situation in real life Bergen, Krivo Flores A Response to Fodor 54 involving actions by simulating—creating ideas of—the perceptual experiences described. The result is that in this view our semantic mental content is not radically distinct from what we feel, perceive, and know how to do but is essentially connected to such capacities Bergen, Predictive projectivism offers the complimentary suggestion that to correctly predict the causal behavior displayed by world is, in large measure, what is necessary to understand that world, and if the embodied simulation hypothesis is on track, such understanding is carried out by the same systems which enable genuine action and perception of the actual world.
Language comprehension is a function of imagining or predicting, albeit often below the level of conscious awareness—an appropriate simulation of situations described. We now are now poised to complete our engagement with Fodor and to summarize our answers to his queries. Krivo Flores 7 Answering Fodor In chapter two I promised to provide answers to two specific critical questions posed by Fodor in Hume Variations, so without further ado: 1. Q: How does the imagination synthesizes token ideas?
Q: How does the imagination know what to imagine? A: the greatest part of the answer the the question of how the imagination knows what to imagine is that, as we have seen, the imagination qua probabilistic generative model—is constantly creating top-down predictions which attempt to match its expected sensory inputs. According to the predictive projectivist model, these same sensory and neural mechanisms serve to co-enact perception, action, imagination, and probabilistic causal inference. The perception of something familiar in the environment, including linguistic stimuli, activates a simulation of the situation predicted, referenced, or described.
Additionally, as Barsalou notes, simulations represent predictions, which can be readily matched to corresponding components, should they occur Barsalou, p. If not, to minimize error, a new conceptualization must be synthesized Barsalou. Krivo Flores A Response to Fodor 56 Another answer is that language is a primary driver of simulation.
This view of Hume is explicitly opposed to recent discussions by critics who hold that the theory of ideas is the Achilles heel of his philosophy and that he would surely have abandoned it if only he had read Wittgenstein carefully. You don't have to know much about Hume to enjoy this inventively argued, provocative, and stimulating defence of the representational theory of mind-which is looking increasingly hard to resist.
Read more Read less. Review Hume Variations relaunches [Fodor's] attack on pragmatism from the perspective of Hume's theory of ideas. No customer reviews. Share your thoughts with other customers. Write a customer review. Discover the best of shopping and entertainment with Amazon Prime. Prime members enjoy FREE Delivery on millions of eligible domestic and international items, in addition to exclusive access to movies, TV shows, and more. Back to top.
Get to Know Us. English Choose a language for shopping. Audible Download Audio Books. Alexa Actionable Analytics for the Web.
Related Hume Variations
Copyright 2019 - All Right Reserved