Transcranial magnetic stimulation to frontal but not occipital cortex disrupts endogenous attention

Abstract

Covert endogenous (voluntary) attention improves visual performance. Human neuroimaging studies suggest that the putative human homolog of macaque FEF (FEF+) is critical for this improvement, whereas early visual areas are not. Yet, MRI methods are correlational, as they do not manipulate brain function. Here we investigated whether rFEF+ or V1/V2 play a causal role in endogenous attention. We used transcranial magnetic stimulation (TMS) to alter activity in visual cortex (Exp.1) or rFEF+ (Exp.2) when observers performed an orientation discrimination task while attention was manipulated. On every trial, they received double-pulse TMS at a predetermined site (stimulated region) around the occipital pole or the rFEF+. Two cortically magnified gratings were presented, one in the stimulated region (contralateral to the stimulated cortical area) and another in the symmetric (ipsilateral) non-stimulated region. Grating contrast was varied to measure contrast response functions (CRFs) for all attention and stimulation combinations. In Exp.1, the CRFs were similar at the stimulated and non-stimulated regions, indicating that early visual areas do not modulate endogenous attention during stimulus presentation. In contrast, occipital TMS eliminates exogenous (involuntary) attention effects on performance (1). In Exp.2, rFEF+ stimulation decreased the overall attentional effect; neither benefits at the attended location nor cost at the unattended location were significant. This pattern is mimicked in the frequency and directionality of microsaccades: Whereas occipital stimulation did not affect microsaccades, rFEF+ stimulation caused a higher microsaccade rate selectively directed toward the stimulated hemifield. These results provide causal evidence of the role of this frontal region for endogenous attention. SIGNIFICANCE STATEMENTHuman neuroimaging studies have revealed activity in frontal regions (e.g., FEF+) as a neural correlate of endogenous (voluntary) attention, and early visual areas (V1/V2) as neural correlates of both endogenous and exogenous (involuntary) attention. Using a causal manipulation-transcranial magnetic stimulation-we show that briefly disrupting activity in rFEF+ weakens endogenous attentions benefits at attended and costs at unattended locations. In contrast, V1/V2 stimulation did not alter endogenous attention (although we have previously demonstrated that it eliminates effects of exogenous attention). Correspondingly, whereas stimulation to rFEF+ increased the rate of microsaccades directed toward the stimulated hemifield, occipital stimulation did not. Together, these results provide causal evidence for the role of rFEF+ but not V1/V2 in endogenous attention.