An adaptive reflexive processing model of neurocognitive function: supporting evidence from a large scale (n = 100) fMRI study of an auditory oddball task.

Recent hemodynamic imaging studies have shown that processing of low probability task-relevant target stimuli (i.e., oddballs) and low probability task-irrelevant novel stimuli elicit widespread activity in diverse, spatially distributed cortical and subcortical systems. The nature of this distributed response supports the model that processing of salient and novel stimuli engages many brain regions regardless of whether said regions were necessary for task performance. However, these latter neuroimaging studies largely employed small sample sizes and fixed-effect analyses, limiting the characterization and inference of the results. The present study addressed these issues by collecting a large sample size (n = 100) and employed random effects statistical models. Analyses were also conducted to determine the inter-subject reliability of the hemodynamic response and the effects of gender and age on target detection and novelty processing. Group data demonstrated highly significant activation in all 34 specified regions of interest for target detection and all 24 specified regions of interest for processing of novel stimuli. Neither age nor gender systematically influenced the results. These data are discussed within the context of a model that proposes that the mammalian brain has evolved to adopt a strategy of engaging distributed neuronal systems when processing salient stimuli despite the low probability that many of these brain regions are required for successful task performance. This process may be termed 'adaptive reflexive processing.' The implications of these results for interpreting functional MRI studies are discussed.