A shared neural substrate for action verbs and observed actions in human posterior parietal cortex

Abstract

High-level sensory and motor cortical areas are activated when processing the meaning of language, but it is unknown whether, and how, words share a neural substrate with corresponding sensorimotor representations. We recorded from single neurons in human posterior parietal cortex (PPC) while participants viewed action verbs and corresponding action videos from multiple views. We find that PPC neurons exhibit a common neural substrate for action verbs and observed actions. Further, videos were encoded with mixtures of invariant and idiosyncratic responses across views. Action verbs elicited selective responses from a fraction of these invariant and idiosyncratic neurons, without preference, thus associating with a statistical sampling of the diverse sensory representations related to the corresponding action concept. Controls indicated that the results are not the product of visual imagery or arbitrary learned associations. Our results suggest that language may activate the consolidated visual experience of the reader.

A parietal region processing numerosity of observed actions: An FMRI study

“…The present finding is in close agreement with our earlier study (Orban et al, 2019) in which phAIP was activated bilaterally when subjects attended to OMAs, rather than to the gender of the actors or color of the targets. These findings, together with the single cell studies mentioned above (Aflalo et al, 2020; Lanzilotto et al, 2019), and a recent stereo‐EEG study (Platonov et al, 2020) support the view that the identity of the OMAs is processed in this region. One difference between the two fMRI studies is that activation in phAIP was observed only in the left hemisphere in the current study.…”

“…These findings are further supported by the observation that featural attention affects selective hemodynamic activity in fMRI (Cukur, Nishimoto, Huth, & Gallant, 2013; Stoppel et al., 2011), as well as by many neurophysiological experiments in non‐human primates (Bisley & Goldberg, 2010; Freedman & Ibos, 2018; Treue & Martinez Trujillo, 1999). Indeed, these latter studies have indicated that featural attention modulates the gain of tuning functions, and recent single cell studies in AIP and its human counterpart have demonstrated neuronal tuning for OMAs (Aflalo et al, 2020; Lanzilotto et al, 2019). It is theoretically true that once action/ball is repeated more than once in the video, participants have sufficient information to solve the task.…”

“…Indeed, different classes of OAs, such as manipulative actions, locomotive actions, or skin displacing actions, are processed in different PPC regions (Abdollahi, Jastorff, & Orban, 2013; Corbo & Orban, 2017; Ferri, Rizzolatti, & Orban, 2015). On the other hand, the exemplars of a given class are processed by OA selective neurons, as has been shown for observed manipulative actions (OMAs) in monkey anterior intraparietal (AIP) area (Lanzilotto et al, 2019) and its human homologue, putative human anterior intraparietal sulcus (phAIP; Aflalo et al, 2020). The link between phAIP and the processing of OMAs identity is further supported by the activation of this region in two‐alternative forced‐choice (2AFC) discrimination tasks with OMAs as discriminanda (Orban, Ferri, & Platonov, 2019; Platonov et al, 2020).…”

Neural encoding of actual and imagined touch within human posterior parietal cortex

“…We recently reported an analogous relation in the parallel domain of motor function ( Aflalo et al, 2020 ; Aflalo et al, 2015 ; Rutishauser et al, 2018 ; Zhang et al, 2017 ). In these studies, we found that a shared PPC neuronal population coded for overt movements as well as cognitive motor variables including imagery, observed actions, and action verbs ( Aflalo et al, 2020 ; Aflalo et al, 2015 ; Andersen and Buneo, 2002 ; Rutishauser et al, 2018 ; Zhang et al, 2017 ). This richness of representation is made possible through a partially mixed encoding in which single neurons represent multiple variables, allowing a relatively small neuronal population (recorded through a 4 × 4 mm implanted microelectrode array) to provide many movement-related signals ( Zhang et al, 2017 ; Zhang et al, 2020 ).…”

“…We recorded from on average 101.6 ± 7.2 neurons ( Figure 1—figure supplement 1 ) over 14 sessions in the PPC (left hemisphere) of a tetraplegic human participant (spinal injury at levels 3–4; C3/4). In previous work, we referred to the implant area as the anterior intraparietal cortex, a region functionally defined in NHPs ( Aflalo et al, 2020 , Aflalo et al, 2015 ; Rutishauser et al, 2018 ; Zhang et al, 2017 ; Zhang et al, 2020 ; Andersen et al, 2019 ; Sakellaridi et al, 2019 ). Here, we refer to the recording site as the PC-IP, acknowledging that further work is necessary to definitively characterize homologies between human and NHP anatomy.…”