The most frequently cited books for applying control theory to human
perception are by J.T. Powers:
Powers, J.T. (1973). Behavior: The control of perception. Chicago: Aldine.
Powers, J.T. (1973). Feedback: Beyond behaviorism. Science, 179, 351-379.
More general references on control theory and control engineering go back at
least 100 years, and you should be able to find good textbook references in
any university bookshop.
I am not familiar with feedforward interviews. This is probably because the
terms 'feedforward' and 'feedback' apply very broadly.
Put simply, control theory defines feedback and feedforward as follows:
To control a process, an operator's task is to keep a process within certain
allowable boundaries (i.e., temperature of water in a large vat equal to 80
Using feedback control, the operator observes the current state of the
system (water at 75 degrees celsius), compares that to the goal state (80
degrees celsius), and performs a control action to current state to the goal
state (heat the water by five degrees). Once you have brought the system to
80 degrees celsius, unpredictable disturbances (like your heater losing
power, or cold water being added to the vat) might cause the temperature to
depart from the goal state. Once you get feedback from the system that this
has happened, you control to compensate. Hence, feedback control.
Now, consider some *predictable* disturbance to the system - for example,
the ambient temperature of the plant is reduced from 21 degrees C to 16
degrees C at 6 p.m. every night. You could control this via feedback
control, and observe that starting at 6 pm the temperature in the vat goes
down because the ambient temperature of the room is going down. You would
use the feedback that the temperature is decreasing to bring the vat back to
its goal state. Or, you could use the fact that you know that the
temperature will decrease predictably at 6 p.m. to pre-emptively begin
adding extra heat to the vat at 6 pm so that the temperature never leaves
the goal state. He you are controlling not based on feedback (a departure
from the goal state because of a disturbance that has already happened) but
on feedforward (a predicted departure from the goal state based on a
disturbance that you know will occur).
When talking about feedforward in the context of the web, this may mean to
give people information as to what the results of a click will be before
that click is made. Giving information as to the results of a click only
after the click has been made is feedback control.
That being said, I still don't know what a feedforward interview is. :-)