Objection 1. It would seem that a habit cannot be corrupted. For habit is within its subject like a second nature; wherefore it is pleasant to act from habit. Now so long as a thing is, its nature is not corrupted. Therefore neither can a habit be corrupted so long as its subject remains.
Objection 2. Further, whenever a form is corrupted, this is due either to corruption of its subject, or to its contrary: thus sickness ceases through corruption of the animal, or through the advent of health. Now science, which is a habit, cannot be lost through corruption of its subject: since "the intellect," which is its subject, "is a substance that is incorruptible" (De Anima i, text. 65). In like manner, neither can it be lost through the action of its contrary: since intelligible species are not contrary to one another (Metaph. vii, text. 52). Therefore the habit of science can nowise be lost.
Objection 3. Further, all corruption results from some movement. But the habit of science, which is in the soul, cannot be corrupted by a direct movement of the soul itself, since the soul is not moved directly. It is, however, moved indirectly through the movement of the body: and yet no bodily change seems capable of corrupting the intelligible species residing in the intellect: since the intellect independently of the body is the proper abode of the species; for which reason it is held that habits are not lost either through old age or through death. Therefore science cannot be corrupted. For the same reason neither can habits of virtue be corrupted, since they also are in the rational soul, and, as the Philosopher declares (Ethic. i, 10), "virtue is more lasting than learning."
On the contrary, The Philosopher says (De Long. et Brev. Vitae ii) that "forgetfulness and deception are the corruption of science." Moreover, by sinning a man loses a habit of virtue: and again, virtues are engendered and corrupted by contrary acts (Ethic. ii, 2).
I answer that, A form is said to be corrupted directly by its contrary; indirectly, through its subject being corrupted. When therefore a habit has a corruptible subject, and a cause that has a contrary, it can be corrupted both ways. This is clearly the case with bodily habits--for instance, health and sickness. But those habits that have an incorruptible subject, cannot be corrupted indirectly. There are, however, some habits which, while residing chiefly in an incorruptible subject, reside nevertheless secondarily in a corruptible subject; such is the habit of science which is chiefly indeed in the "possible" intellect, but secondarily in the sensitive powers of apprehension, as stated above (50, 3, ad 3). Consequently the habit of science cannot be corrupted indirectly, on the part of the "possible" intellect, but only on the part of the lower sensitive powers.
We must therefore inquire whether habits of this kind can be corrupted directly. If then there be a habit having a contrary, either on the part of itself or on the part of its cause, it can be corrupted directly: but if it has no contrary, it cannot be corrupted directly. Now it is evident that an intelligible species residing in the "possible" intellect, has no contrary; nor can the active intellect, which is the cause of that species, have a contrary. Wherefore if in the "possible" intellect there be a habit caused immediately by the active intellect, such a habit is incorruptible both directly and indirectly. Such are the habits of the first principles, both speculative and practical, which cannot be corrupted by any forgetfulness or deception whatever: even as the Philosopher says about prudence (Ethic. vi, 5) that "it cannot be lost by being forgotten." There is, however, in the "possible" intellect a habit caused by the reason, to wit, the habit of conclusions, which is called science, to the cause of which something may be contrary in two ways. First, on the part of those very propositions which are the starting point of the reason: for the assertion "Good is not good" is contrary to the assertion "Good is good" (Peri Herm. ii). Secondly, on the part of the process of reasoning; forasmuch as a sophistical syllogism is contrary to a dialectic or demonstrative syllogism. Wherefore it is clear that a false reason can corrupt the habit of a true opinion or even of science. Hence the Philosopher, as stated above, says that "deception is the corruption of science." As to virtues, some of them are intellectual, residing in reason itself, as stated in Ethic. vi, 1: and to these applies what we have said of science and opinion. Some, however, viz. the moral virtues, are in the appetitive part of the soul; and the same may be said of the contrary vices. Now the habits of the appetitive part are caused therein because it is natural to it to be moved by the reason. Therefore a habit either of virtue or of vice, may be corrupted by a judgment of reason, whenever its motion is contrary to such vice or virtue, whether through ignorance, passion or deliberate choice.
Reply to Objection 1. As stated in Ethic. vii, 10, a habit is like a second nature, and yet it falls short of it. And so it is that while the nature of a thing cannot in any way be taken away from a thing, a habit is removed, though with difficulty.
Reply to Objection 2. Although there is no contrary to intelligible species, yet there can be a contrary to assertions and to the process of reason, as stated above.
Reply to Objection 3. Science is not taken away by movement of the body, if we consider the root itself of the habit, but only as it may prove an obstacle to the act of science; in so far as the intellect, in its act, has need of the sensitive powers, which are impeded by corporal transmutation. But the intellectual movement of the reason can corrupt the habit of science, even as regards the very root of the habit. In like manner a habit of virtue can be corrupted. Nevertheless when it is said that "virtue is more lasting than learning," this must be understood in respect, not of the subject or cause, but of the act: because the use of virtue continues through the whole of life, whereas the use of learning does not.
Source: Summa Theologica (New Advent)