ABSTRACT
For nearly 100 years, intelligence theorists have debated whether cognitive ability is best conceived as a molar or modular construct. A molar system is one in which a unitary, general process functions across a wide variety of cognitive tasks. A modular system involves numerous distinct cognitive processing units each responsible for certain nonoverlapping cognitive tasks. Modularity is currently the dominant theory in developmental cognitive psychology and neuroscience (e.g., Karmiloff-Smith, 1992). In this chapter, I argue that theories of modularity are based primarily on normative designs or studies of clinical populations, and thus may not apply to variance in the cognitive functioning of unselected populations. More importantly, I argue that theories of modularity have not taken important behavioral genetic results into account. Specifically, behavioral genetic results point strongly to a general factor that becomes more important and more genetic across the lifespan.
