Even when we know what a man thinks it his interest to do, we do not know for certain what he will do.

The man who studies the Stock Exchange list does not buy his Debentures, unless, apart from his intellectual inference on the subject, he has an impulse to write to his stockbroker sufficiently strong to overcome another impulse to put the whole thing off till the next day.
Macaulay might even further have admitted that the mental act of calculation itself results from, or is accompanied by, an impulse to calculate, which impulse may have nothing to do with any anterior consideration of means and ends, and may vary from the half-conscious yielding to a train of reverie up to the obstinate driving of a tired brain onto the difficult task of exact thought.
The text-books of psychology now warn every student against the 'intellectualist' fallacy which is illustrated by my quotation from Macaulay.

Impulse, it is now agreed, has an evolutionary history of its own earlier than the history of those intellectual processes by which it is often directed and modified.

Our inherited organisation inclines us to re-act in certain ways to certain stimuli because such reactions have been useful in the past in preserving our species.

Some of the reactions are what we call specifically 'instincts,' that is to say, impulses towards definite acts or series of acts, independent of any conscious anticipation of their probable effects.[4] Those instincts are sometimes unconscious and involuntary; and sometimes, in the case of ourselves and apparently of other higher animals, they are conscious and voluntary.
But the connection between means and ends which they exhibit is the result not of any contrivance by the actor, but of the survival, in the past, of the 'fittest' of many varying tendencies to act.