I think the PET guys that spent a lot of time on progressive subtraction of paradigms might have good perspective on this. Marcus Raichle wrote about differences between resting state data when passively viewing a cross hair vs eyes closed.

The brain does so much covert processing, your answer will ultimately depend on the level of inference you wish to draw.  You could always test it like Raichle et al. did.

Raichle M et al. A default mode of brain function (2001) PNAS 98(2) 676-682. http://www.pnas.org/content/98/2/676.abstract

See the last two paragraphs of results section; they say more details to be published separately, but I never saw anything. I'd be interested if you track down further details.

Hope this helps,

Reese

Given that the whole brain is active, all the time, it is neccessary to do some subtraction analysis to make sense of the data, and to relate the brain activation to cognitive processes. That being said, I agree that you might miss out on valuably information when doing a subtraction. To echo Reese, I guess it depends on what type of inferences you would like to draw. If you want to investigate a very specific frontal lobe function, say inhibition, it might make sense to subtract away other influences, for example neutral sensory input or some task does not rely on inhibition. As Reese mentioned, resting state networks will give you an idea of frontal lobe activation without the need to do subtraction. Several of these resting state networks heavily involve the frontal lobe, e.g. default mode network, executive network. Just my thoughts, but I have no specific articles which would be of interest to you.

Thanks, all.  I ended up conducting analyses both ways.  Hopefully someone will specifically investigate the neutral cue impact soon.  Some food for thought:

-Neutral cue response is modulated during pharmacological treatment

Outhred T, Das P, Felmingham KL, Bryant RA, Nathan PJ, Malhi GS, Kemp AH. Impact of acute administration of escitalopram on the processing of emotional and neutral images: a randomized crossover fMRI study of healthy women. Journal Of Psychiatry & Neuroscience: JPN 2014;39 (4), 267-275.