What is presented as "Pavlov's Dog" is a misinterpretation (by Pavlov, who should have stuck to merringue-based desserts, and others) of certain experiments. In effect, Pavlov's experiments are put forward as demonstrating the truth of an associationist psychology a la Hume and Locke. But the experiments permit a non- or anti-associationist interpretation, and therefore the experiments do not demonstrate which of the two interpretations is correct. A better way into this, perhaps, is not Pavlov's work, which permits rival interpretations, but the work of John Garcia which undermines the associationist interpretation. In Garcia's experiments, as explained by Martin Seligman:- "Every time his rats licked at their drinking spouts, they tasted saccharin, and a burst of light and noise came on. This is called a _compound CS_ - bright, noisy saccharin water. Then a burst of X rays occurred. Within a few hours the rats were sick to their stomachs. When the rats recovered, Garcia tested the elements of the compound CS separately to see what the rats had learned to fear. They now hated saccharin, but they were completely unperturbed by the bright noise. When they got sick, they blamed it on the taste and ignored everything else. "Maybe they had just failed to notice the bright noise duing conditioning? So Garcia counterbalanced the experiment. Other rats were given the same bright, noisy saccharin paired now with foot shock instead of stomach illness. What did they learn? They now cringed in fear of bright noise, but they still loved saccharin. When they suffered pain, they blamed it on the bright noise and ignored the taste. "So the bright noise and the saccharin were noticeable, but only the taste became aversive when the rats became nauseated, and only the bright noise became aversive when the rats suffered pain. How could this be?" Seligman points out difficulties in fitting this within the framework of Pavlovian conditioning ['PC']. These include:- PC is not selective whereas the rats are selective as to what they associate with what; PC depends on short time gaps whereas the rats 'associated' "saccharin" and the nausea they felt hours later; PC depends on repetition whereas the rats were put off the food after a single episode of post-food nausea; PC is meant to involve 'awareness' but rats nauseated while anaesthetised woke up with an aversion to saccharin; PC is meant to be as easily extinguished but this is not true of the rat's 'learned' aversions. Seligman concludes, however, that all these difficulties "_can_ be reconciled with Pavlovian conditioning, because evolution is at work". Be that as it may, Popper argues that PC, and its arc of the conditioned/conditional reflex, is a misinterpretation that cannot be easily reconciled with a Darwinian view of evolution: instead it is a Lamarckian and an inductivist misinterpretation. It might be tempting (particularly to those pre-conditioned or predisposed that way) to see the rats' aversive reactions as a special subset of Pavlovian conditioning that have arisen given their usefulness as a survival mechanism (survival mechanisms for avoiding poisonous foods or electrical shocks). But for a thinker like Popper the correct conclusion is that Garcia's rats give us the better clue to the general picture, not Pavlov's dogs; and that Pavlov's dogs reactions are a special subset of the model of Darwinian trial-and-error learning, not associationist learning, that better accounts for the rats' and the dogs' behaviour. Absent poisonous after-effects, which it would be evolutionarily useful to learn from, and bearing in mind the gradual winnowing from the gene-pool of creatures that are over-apt to eat poisonous things (a winnowing so successful that other creatures can take advantage and avoid being eaten by dissembling that they are poisonous or harmful to eat):- creatures like dogs and rats would be well-advised to eat whenever the opportunity arises and will do so provided prior disposition or past experience does not cast the food as harmful. As dogs need to salivate to help their eating, evolution will favour a great readiness to salivate in order to that the dog be able to take advantage of any eating opportunity: this predisposition to salivate at the slightest expectation of an eating opportunity explains why dogs will readily salivate if the person who normally feeds them turns up or if it is normal 'feeding time' etc. _This predisposition is not the result of PC rather what looks like PC is a result of this predisposition_. The same is true of the rats. They are not conditioned as per PC: rather they are evolutionarily prepared both to eat and to avoid poison, and evolutionarily prepared with mechanisms for differentiating what will nourish them and what will poison them - one such mechanism is a learning-mechanism where they 'learn' from post-feed nausea to avoid the kind of food in question. This learning is not associationist in the sense of PC but is better understood as an imperfect 'trial-and-error' mechanism that has evolved by natural selection. Donal Enjoying Mozart London ------------------------------------------------------------------ To change your Lit-Ideas settings (subscribe/unsub, vacation on/off, digest on/off), visit www.andreas.com/faq-lit-ideas.html