In Jon Schmidt's excellent article "Contemplating Competence" in EOR Viewpoint (11/18/11), he describes how competence is based on skill acquisition, not attainment of information or classroom learning.   Below he describes skill acquisition based on Drefus's phenomenological study:

Philosopher Hubert Dreyfus and his brother Stuart, an industrial engineer, conducted a phenomenological study of various "unstructured" problem areas, such as task environments that contain a potentially unlimited number of details that may or may not be pertinent. The resulting model of skill acquisition:

1. The "novice" complies with strict rules based on context-free features of the task environment.

2. The "advanced beginner" recognizes situational aspects of the task environment and follows maxims to adjust his or her actions accordingly.

3. The "competent performer" does not try to account for all discrete elements of the task environment; instead, he or she selects a plan, goal or perspective to establish which elements are relevant and which may be safely ignored.

4. The "proficient performer" no longer reflects on the task environment as a detached observer; without having to evaluate multiple options, he or she simply sees what needs to be done and then chooses how to go about doing it.

5. The "expert" intuitively perceives both what needs to be done and how to do it, making extremely subtle and refined discriminations in a variety of task environments that are sufficiently similar to those previously encountered. [Schmidt]

The higher levels are attained when one becomes an engineer, and that requires extensive experience working in the industry.   Jon  suggests these should become the standards for licensure (and perhaps they should).  But I think what separates the competent from the expert is the ability to intuitively understand the problem, identify the solutions and decide on the best solution for the problem with little need to exhaustively analyze each possible solution.  How can a test be structured to account for this design ability?  (for another topic)  Currently, in the United States, one is able to practice enigneering after four years of experience and passing two tests (EIT and PE test).

He goes on to say:

However, when confronted with an unfamiliar set of circumstances, even experts inevitably revert to the behaviors of novices and advanced beginners. They must fall back on rules and maxims because they lack the kind or amount of experience that would enable them to discern the appropriate course of action. [Schmidt]

So, engineering does not require intellegence in the normal sense of the word.   It requires competence and proficiency.  The best of us, the experts, are those who work intuitively and able to make refined discriminations based on past successes and failures in order to make the best design decisions.   This is a combination of natural ability, education, and experience (I guess that is obvious).