24/09/2026
This is particularly interesting from an ERP perspective. With ERPs we can identify different stages of information processing with very high temporal precision — essentially asking not only *whether* the brain responds, but *when* the processing begins to diverge.
In an auditory oddball paradigm, for example, we can examine early sensory processing and discrimination, attentional orienting and novelty detection, and later processes involved in evaluating and updating the significance of a stimulus. Components such as N1, P2, mismatch-related responses, P3a and P3b can help us distinguish where in this processing sequence an atypical response occurs.
The important distinction is that fMRI and ERPs provide different kinds of information. fMRI is much better at identifying the network of anatomical regions recruited during novelty processing. ERPs provide millisecond-level information about the sequence and dynamics of information processing, but conventional scalp ERPs cannot tell us with the same confidence that, for example, 13 rather than 23 specific anatomical regions were activated.
Source-localisation methods can estimate the cortical generators contributing to an ERP, but this remains an inverse problem and should not be interpreted as equivalent to fMRI localisation.
So the two approaches are highly complementary: fMRI tells us more about *where* the novelty network is recruited, while ERPs can tell us much more precisely *when* processing diverges — sensory registration, novelty detection, attentional orienting, or later context updating.
For ADHD, this can be much more informative than simply saying that someone has an “attention deficit.” The processing difficulty may occur at a particular stage, and it may differ substantially between individuals.
ADHD may not simply be a problem of “not paying attention.” An fMRI study of adolescents with ADHD showed something more specific. Their brains could activate many of the same regions as healthy adolescents when they were asked to detect important target sounds.
But when an unexpected, novel sound appeared, the difference became much more noticeable.
The ADHD group activated only 13 of 31 brain regions involved in novelty processing, compared with 23 of 31 regions in the control group. These regions included areas involved in attention, working memory, and orienting toward something new.
So the interesting question may not be simply, “Can the brain pay attention?”
It may be: “How efficiently does the brain notice and shift toward something that suddenly becomes important?”
The study also found slower and more variable performance in the ADHD group, supporting the idea that attentional processing can be inconsistent rather than simply absent.
One important limitation: this was a relatively small study of boys with combined-type ADHD, so these findings should not be generalized to everyone with ADHD.
Reference: Stevens MC, Pearlson GD, Kiehl KA. An fMRI Auditory Oddball Study of Combined-Subtype Attention Deficit Hyperactivity Disorder. American Journal of Psychiatry. 2007
