Intracranial Recordings Demonstrate Both Cortical and Medial Temporal Lobe Engagement in Visual Search in Humans

Abstract

Visual search is a fundamental human behavior, providing a gateway to understanding other sensory domains as well as the role of search in higher-order cognition. Search has been proposed to include two component processes: inefficient search (search) and efficient search (pop-out). According to extant research, these two processes map onto two separable neural systems located in the frontal and parietal association cortices. In this study, we use intracranial recordings from 23 participants to delineate the neural correlates of search and pop-out with an unprecedented combination of spatiotemporal resolution and coverage across cortical and subcortical structures. First, we demonstrate a role for the medial temporal lobe in visual search, on par with engagement in frontal and parietal association cortex. Second, we show a gradient of increasing engagement over anatomical space from dorsal to ventral lateral frontal cortex. Third, we confirm previous intracranial work demonstrating nearly complete overlap in neural engagement across cortical regions in search and pop-out. We further demonstrate pop-out selectivity, manifesting as activity increase in pop-out as compared to search, in a distributed set of sites including frontal cortex. This result is at odds with the view that pop-out is implemented in low-level visual cortex or parietal cortex alone. Finally, we affirm a central role for the right lateral frontal cortex in search.

Asymmetric coding of reward prediction errors in human insula and dorsomedial prefrontal cortex

“…However, this study reported region-level effects, leaving the diversity in local population coding of RPE salience and valence unexplored. This emphasizes the need to disentangle spatially intermingled circuits performing different computations within a given region, which is characteristic of previous human iEEG findings in language and attention 57,58 . Furthermore, we found that the proportions of channel sites coding RPE salience, RPE value, and positive and negative RPEs were equivalent in dMPFC and INS.…”

Visual Attention and Poor Sleep Quality

“…Despite not explicitly and overtly implementing a serial visual search task, we believe that this finding probably reflects a higher load on the DAN in PSs and not using the VAN by this group for visual search. Also, since it seems that the serial and parallel search strategies are more reliant on the DAN and VAN, respectively ( Wang et al, 2020 ; Ischebeck et al, 2021 ; Slama et al, 2021 ), this finding indicates employing the serial visual search strategy in PSs. Importantly, the exact localization of the TPJ as an important node of the VAN is controversial; however, the supramarginal gyrus is considered a major node of TPJ ( Uddin, 2015 ; Bernard et al, 2020 ).…”

“…By contrast, the right supramarginal gyrus, a major node in the VAN, was activated by processing the targets, probably indicating the template-matching function in this region ( Macaluso and Ogawa, 2018 ). Direct intracranial recordings also support the notion that parallel search strategies are reliant on the VAN ( Slama et al, 2021 ). Moreover, inactivating DAN nodes (FEFs and IPS) using repetitive transcranial magnetic stimulation disrupted the non-pop-out search strategy, while it had no effect on the pop-out visual search, proving a causal role of the DAN in the serial visual search ( Wang et al, 2020 ).…”

A functional overlap between evidence accumulation, confidence, and changes of mind in the pre-supplementary motor area and insula

“…Indeed results from single-unit recordings in non-human primates show that neuronal activity is very heterogeneous even within a small area such as in the lateral intraparietal area So & Shadlen, 2022). Similarly, spatial selectivity in sEEG was found for language or attention (Flinker et al, 2011;Slama et al, 2021). This possibility would explain the diversity of results observed in the literature when comparing imaging techniques with lower spatial resolution such as EEG, and works using high spatial resolution techniques such as single-unit recordings (Kiani et al, 2008;Morito & Murata, 2022;van Vugt et al, 2012).…”