The current trade-off account builds on the ideas of
MacDonald (1995), who argued that the observed range of
variation represents the range of viable human behavioral
strategies and who stressed that there are fitness disadvantages
at the extremes. Thus, he stressed stabilizing selection.
The present argument is that selection can fluctuate,
such that it may sometimes be directional for increasing a
trait and sometimes be directional for decreasing it. Among
the great tits, for example, selection on exploration is
clearly directional in any given year (Dingemanse et al.,
2004). The retention of a normal distribution is a consequence
of the inconsistency of the direction of selection,
not its stabilizing form. That said, I agree with MacDonald
that there could be quite general disadvantages at the extremes
of some personality dimensions, such as chronic
depression with high neuroticism, or obsessive– compulsive
personality disorder with high conscientiousness. It is
not a necessary feature of the current approach that there
always be stabilizing effects