The Functional Neuroanatomy of Target Detection: An fMRI Study of Visual and Auditory Oddball Tasks

Abstract

The neuronal response patterns that are required for an adequate behavioural reaction to subjectively relevant changes in the environment are commonly studied by means of oddball paradigms, in which occasional 'target' stimuli have to be detected in a train of frequent 'non-target' stimuli. The detection of such task-relevant stimuli is accompanied by a parietocentral positive component of the event-related potential, the P300. We performed EEG recordings of visual and auditory event-related potentials and functional magnetic resonance imaging (fMRI) when healthy subjects performed an oddball task. Significant increases in fMRI signal for target versus non-target conditions were observed in the supramarginal gyrus, frontal operculum and insular cortex bilaterally, and in further circumscribed parietal and frontal regions. These effects were consistent over various stimulation and response modalities and can be regarded as specific for target detection in both the auditory and the visual modality. These results therefore contribute to the understanding of the target detection network in human cerebral cortex and impose constraints on attempts at localizing the neuronal P300 generator. This is of importance both from a neurobiological perspective and because of the widespread application of the physiological correlates of target detection in clinical P300 studies.

The hemodynamics of oddball processing during single-tone and two-tone target detection tasks

“…These generators are believed to be located in bilateral middle and superior temporal gyrus, and bilateral inferior and middle frontal gyrus (Anderer et al, 1998;Tarkka et al, 1995;Yamazaki et al, 2000). A link between hemodynamic activity in these brain areas and the N1 ERP also are found in several recent studies that have tried to link hemodynamics (Horovitz et al, 2002;Liebenthal et al, 2003;Linden et al, 1999;Menon et al, 1997;Opitz et al, 1999) or event-related magnetic fields (Siedenberg et al, 1996) to the early ERP components observed in oddball paradigms. A simultaneous EEG/MEG study found that N1 waveforms during two-tone auditory oddball rare target detection was similar in both ERP and MEG, and were localized to temporal lobe regions (Siedenberg et al, 1996).…”

“…Opitz et al (1999) found that hemodynamic activity in bilateral temporal gyri was correlated with the amplitude of the N1 elicited during a mismatch negativity (MMN) paradigms for both attended and unattended deviant stimuli. Linden et al (1999) interpreted hemodynamic activity in the temporal lobe as relating to the process of updating memory of the auditory input sequence, and possibly related to ERP components that occur earlier than the P3.…”

“…ERP source localization studies suggest that the P3 is localized to the bilateral prefrontal cortex, parietal-occipital junction, and parietal lobe, while the N1 is localized predominantly to bilateral auditory cortex, with some evidence supporting additional sources in the prefrontal cortex (Anderer et al, 1998;Giard et al, 1994;Halgren et al, 1995a;Liegeois-Chauvel et al, 1994). Functional magnetic resonance imaging (fMRI) studies have confirmed and extended these findings Casey et al, 2001;Clark et al, 2001;Clark et al, 2000;Kiehl et al, 2001a;Kiehl et al, 2001b;Kiehl & Liddle, 2003;Kiehl et al, 2005;Kirino et al, 2000;Linden et al, 1999;McCarthy et al, 1997;Menon et al, 1997;Opitz et al, 1999;Stevens et al, in press;Strange & Dolan, 2001). These studies report that target detection elicits widespread cortical and subcortical activation, including multiple foci in the prefrontal, temporal, parietal and occipital lobes, and subcortical sites.…”

“…Opitz et al (1999) found that hemodynamic activity in bilateral temporal gyri was correlated with the amplitude of the N1 elicited during a mismatch negativity (MMN) paradigms for both attended and unattended deviant stimuli. Linden et al (1999) interpreted hemodynamic activity in the temporal lobe as relating to the process of updating memory of the auditory input sequence, and possibly related to ERP components that occur earlier than the P3.Because several ERP studies have found no differences in the amplitude of the P3 in these conditions, it is unlikely that differences in the amplitude of the hemodynamic response between single-and two-tone oddball performance can be explained within the context of working memory and contextual updating cognitive processes commonly offered as the interpretation of oddball-task P3 results (Donchin & Coles, 1988). Therefore, the most parsimonious cognitive interpretation of the differences in the amplitude of the hemodynamic response between conditions in this study is that they likely are related to cognitive processes proposed to describe the early and middle-latency potentials N1, P2 and N2.…”

Auditory and visual connectivity gradients in frontoparietal cortex

“…However, it is unclear whether the same frontoparietal regions are recruited for both auditory and visual tasks. Several studies have reported activity in a dorsal frontoparietal network that overlaps with the visual system during auditory stimulus anticipation, processing and attention shifts [Bharadwaj et al, 2014; Downar et al, 2000; Krumbholz et al, 2009; Langner et al, 2011; Lee et al, 2012; Linden et al, 1999; Mayer et al, 2006; Shomstein and Yantis, 2004; Watkins et al, 2007; Wu et al, 2007]. In contrast, there is increasing evidence that visual and auditory attention are subserved by different networks [Braga et al, 2013b; Bushara et al, 1999; Degerman et al, 2006; Kong et al, 2014; Maeder et al, 2001; Salmi et al, 2007; Seydell‐Greenwald et al, 2014].…”