Mental Toughness

One thing James Smith has preached a lot recently is the elimination of the current construct of "mental toughness." It is non-specific and does not provide context or quantification. What he stresses is that this "mental toughness" is actually a collection of several qualities, including aggressiveness, durability/resilience, resolve, composure, and positive thinking, among others.
These qualities are all trainable, and in particular, within the sport. He also notes that, "there is no such thing as tough; you are either trained or untrained." Our response in a given situation is a result of life experiences paired with inborn genetic traits. This is training. Consider the sport of American football. Perhaps, for instance, you have a defensive player who is unwilling to be involved in tackling beyond cursory attempts. They will not "stick their nose in." This refusal is generally deemed that the player is not "tough" enough. What they lack is a desire to engage in contact, colored by their past experiences, or, void of those experiences (such as someone playing for the first time), what they anticipate will be the experience upon contact.
Coaches must realize the ability they have to control and improve this. The immediate reaction is likely to vilify, demean, or shame the individual; and/or simply run them until they puke. Setting aside the absurdity of physical punishment, what would be wrong with simply reducing the intensity of the "exercise" (in this instance, tackling), via a reduction in speed and hitting power? Reinforce proper takedown mechanics, repeat as much as necessary, producing an "inoculation" against the stressor (the experience of the tackle). Once this has taken hold, gradually increase the intensity, such that they become immune to the negative experience of the tackle itself. Certainly there are those with inborn traits which predispose them to actually enjoying this experience - these athletes, just like any who have genetic advantages in certain abilities, will not have to necessarily start with the regressed versions.
Analyze the athlete, work backwards from the goal, and construct a logical plan to get there.