To appreciate the field of artificial intelligence, it is helpful to understand that it isbeing pursued along two paths. One is the engineering track in which researchersare trying to develop systems that exhibit intelligent behavior. The other is a theoreticaltrack in which researchers are trying to develop a computational understandingof animal—especially human—intelligence. This dichotomy is clarifiedby considering the manner in which the two tracks are pursued. The engineeringapproach leads to a performance-oriented methodology because the underlyinggoal is to produce a product that meets certain performance goals. The theoreticalapproach leads to a simulation-oriented methodology because the underlyinggoal is to expand our understanding of intelligence and thus the emphasis is onthe underlying process rather than the exterior performance.