Retinotopically specific effects of attention on human early visual cortex activity.

Abstract

Attention helps us to select what is relevant from the enormous amounts of information taken up by our senses. However, it remains unclear just how early in sensory processing attentional selection can occur. Here, we investigated this question in healthy volunteers by assessing the effect of attentional load on the earliest component (C1) of the visual evoked potential (VEP). We mapped participants’ C1 responses to task-irrelevant peripheral textures of different densities and then selected those textures eliciting maximal C1 amplitudes in the upper and lower visual field in each participant. In a second experimental session, these optimal C1 stimuli served as peripheral distracters while participants performed easy or difficult detection tasks at fixation. Our results show a reduction of C1 amplitudes under high attentional load selectively in the lower visual field. This asymmetric effect is opposite to previously reported results obtained without preselecting stimuli. We conclude that attentional selection during early visual processing is possible, but depends on the interaction between anatomic and functional anisotropies of the visual system. This underscores the importance of precisely delineating when, where, and how attentional filtering can operate on initial perceptual processing.

A systematic exploration of attentional load effects on the C1 ERP component

“…Moreover, regardless of the specific experimental design used, the subsequent extrastriate P1 component was not modulated by attentional load either. These ERP results are compatible with some earlier ERP studies (Ding et al, 2014; Herde et al, 2022; Rauss et al, 2009, 2012; Wolf et al, 2022), but not with other ones (Fu et al, 2010b; Rossi & Pourtois, 2012). According to the load theory of selective attention (Lavie, 1995, 2005; Lavie et al, 2004; Lavie & Tsal, 1994), under high load, the filtering or suppression of task‐irrelevant distractors is facilitated compared to low load, and this attention effect is deemed perceptual , influencing early stages of visual processing.…”

“…Besides the lack of C1 modulation by attentional load, we also found that the subsequent extrastriate P1 component was not affected by this factor either. This latter result is not surprising as many previous ERP studies on attentional load (Ding et al, 2014;Herde et al, 2022;Rauss et al, 2009Rauss et al, , 2012Wolf et al, 2022) already reported a lack of P1 modulation (but see Fu et al, 2010b for an exception). In comparison, several previous ERP studies already reported P1 modulations as a function of spatial attention (see for example Baumgartner et al, 2018;Di Russo et al, 2003;Martínez et al, 1999).…”

“…Some slight changes in the stimulus or task parameters might also have a profound influence on the C1, and eventually its modulation by attention (Fu, 2018; Herde et al, 2020). For example, a recent ERP study showed that increasing attentional load at fixation reduced the C1 elicited by the peripheral distractors in the LVF but not in the UVF (Herde et al, 2022). This is opposite to what has been found in several previous studies where modulation of the C1 was mostly found in the UVF (Pourtois et al, 2008; Rauss et al, 2009, 2012).…”

“…A methodological factor that has not been considered yet in these existing studies is the length or duration of the stimulus onset asynchrony (SOA) between the central stimulus and the peripheral distractor. In previous ERP studies on attentional load, a gap was usually introduced between the central task‐relevant stimulus and the peripheral task‐irrelevant distractor (i.e., SOA of 500–750 ms in Herde et al, 2022; SOA of 500–743 ms in Rauss et al, 2009; SOA of 500–750 ms in Rossi & Pourtois, 2012, 2017). Although the SOA varied in these earlier studies, they did not directly compare short to long SOAs.…”

“…Usually, changes in perceptual load are achieved by varying the amount of stimuli (or distractors) to be processed at the same time (Lavie, 1995;Lavie & Tsal, 1994). In comparison, changes in attentional load are obtained by keeping the visual input constant between conditions, but varying task demands based on it (Brockhoff et al, 2022;Wolf et al, 2022), for example, by asking participants to perform either simple detection or conjunction search (Herde et al, 2022;Rauss et al, 2009Rauss et al, , 2012, or passive viewing versus active task (Fu et al, 2010b). Hence, although perceptual load and attentional load likely share some common ground (i.e., more attention resources are allocated in the high than low load condition), they can be dissociated from each other at the methodological level.…”