The author discusses symptoms which are "learned" from certains somatic diseases. When the disease is extinguished such symptoms persist "without a cause", or they may easily recur. Small insults to the homeostasis bring back the "learned" or "engraved" symptoms. They are not conditioned reflexes since in these there is a specific conditioning stimulus while in the cases here mentioned the insults to the homeostasis are not specific and of a varied nature, either psychical or somatic. The cases reported presented several different clinical pictures: effort syndrome, allergic rhinitis, hyperphagia, pseudo-hyperthyroidism, irritable colon, peptic ulcer, migraine. A very clear psychogenic triggering mechanism has produced mistakes and induced the thought of a psychogenic etiology when there are not only a psychogenic but also somatogenic triggering mechanisms and not "causes". Special emphasis was given to the inespecific factors of the crises and not to "specific psychogenic causes" of the disease. The presence of a sense of symptoms so hardly looke for by psychotherapists nowadays was studied. The finality value of the sense of the symptoms may be secondary and does not invalidate the medico-legal legitimacy of them. The gradual facilitation of the symptoms comes to the point of constituting a very peculiar state which the author calls "personal preferential reaction" since any alteration of the homeostasis will invariably produces the same symptoms. As a consequence, their "psychicalization" becomes very easy, that is, the somatic symptoms develop deep bonds with the mental equilibrium and any alteration in this will bring out preferential symptoms. Despite the bonds between them and emotional stability, the symptoms have no finality at least in the beginning, before "psychicalization". The engraving of preferential functional pathways is a notion which has been registered by neurophysiologists; in the present paper it was applied to clinical problems, widening the chapter of the "choice of symptoms". Old problems which have puzzled those interested in psychosomatic aspects of disease were alluded to. An example of such problem is the fact that only inespecific factors are found in migraine, peptic ulcer, colitis, etc. instead of specific causes. An answer was provided: the imprinting of preferential functional pathways is activated by any alteration in the homeostasis and the organism has a "personal preferential reaction" which is always the same.