Neural markers of automatic and controlled attention during immediate and delayed action

Abstract

It has been shown that visually guided and memory-guided actions are under the control of dissociable neural systems. This experiment measured event-related potentials (ERPs) in a cross-modal dual-task paradigm to investigate the attentional requirements of these two types of actions. In a primary joystick-controlled continuous reciprocal aiming task, participants moved a cursor back and forth between two targets of variable size in visually guided (VIS) and memory-guided (MEM) conditions. In a secondary dichotic listening task, ERPs were collected while infrequent high and frequent low pitch tones were delivered to both ears. Participants responded to the infrequent tones delivered to only one attended ear. Aiming and listening were performed separately and together as a dual task. We were interested in two ERP components: the P300 component, which reflects voluntary attention, and the mismatch negativity (MMN), which reflects automatic attention. The results showed that the P300 component elicited by the auditory task was decreased in amplitude by the dual-task conditions compared with the auditory task alone. Moreover, P300 latency was increased by the MEM aiming condition, but not the VIS aiming condition. On the other hand, the MMN component elicited by the auditory task was only attenuated by the VIS aiming condition, not by the MEM aiming condition. Together, these results suggest that memory-guided aiming requires more voluntary attention and less automatic attention than visually guided aiming.

Theta oscillations reflect a putative neural mechanism for human sensorimotor integration

“…Likewise, the parietal region did not display differences in the amount of theta activity between conditions, consistent with evidence that both types of reaching rely on the dorsal stream (Franz et al 2009). During the initiation phase, theta activity was greater for condition 2 at the left temporal region (located near ventral stream areas), corroborating behavioral, neuropsychological, and neuroimaging reports that perceptual brain mechanisms in the ventral stream are recruited when planning perceptually driven hand actions (Armstrong and Singhal 2011;Cohen et al 2009;Goodale et al 1991;James et al 2003;Milner et al 2001;Singhal et al 2006). Thus, as the sensorimotor network shifts to include ventral regions, theta activity increases at the ventral site, during the phase when perceptual mechanisms are recruited.…”

“…These types of actions still engage dorsal structures (Murata et al 1996), but according to the influential perception-action model of Goodale and Milner (1992), they also critically depend on perceptual mechanisms residing in the ventral stream. Behavioral and neuroimaging evidence suggests that even when occlusion of a target coincides with a response cue, or occurs shortly beforehand, ventral stream mechanisms are necessarily recruited (Armstrong and Singhal 2011;Singhal et al 2006Singhal et al , 2007Westwood and Goodale 2003). In the current study, the target disappeared at either the end of the participants' reaction time (condition 1) or the beginning (condition 2), thereby altering the contribution of the ventral stream at the response initiation phase.…”

“…1). Similar timing has been used previously (Armstrong and Singhal 2011;Singhal et al 2007;Westwood and Goodale 2003), and in our task the target disappeared in both conditions so that trials were made as similar as possible. By manipulating whether the target was available on the retina during movement onset, we were able to manipulate the network required to perform the task in a way that was imperceptible to participants (Singhal et al 2007;Westwood and Goodale 2003).…”

Human electrophysiological reflections of the recruitment of perceptual processing during actions that engage memory

“…That is, MT was increased and reach accuracy was decreased in condition 2, which compared to condition 1, engages ventral mechanisms to a greater extent. This pattern has been shown in other studies comparing visually guided actions and those for which vision of a target is precluded (Armstrong & Singhal, 2011;Klatzky et al, 1993;Singhal et al, 2007;Westwood & Goodale, 2003); tasks which similarly differ in their reliance on the ventral stream. These findings support the idea that actions that engage the ventral stream rely on stored perceptual information, which is less accurate than real-time visual information and thus, performance to visually unavailable targets induces slowing and greater variability in the arm and hand actions (Goodale & Milner, 1992 (post hoc) that they were unaware of which trial type they were engaged in, suggesting that our manipulation was successful in engaging perception and memory processes without altering performance strategy.…”

Emotion matters: Implications for distracted driving

“…One possible explanation for the lack of difference in recall between single-and dual-task conditions is that two forms of attention may be utilized during the driving and recall task. The literature strongly suggests that there are two forms of attention: bottom-up automatic attention and top-down controlled attention (e.g., Armstrong and Singhal, 2011). It is possible that automatic processes that require little executive attention may be associated with the driving task, while controlled processes that rely on executive control (e.g., working memory) may be associated with the encoding and storage of items in memory.…”