Top-down control in selective attention

Project leader

Funding source

Swedish Research Council - Vetenskapsrådet (VR)

Project Details

Start date: 01/01/2016
End date: 31/12/2019
Funding: 4000000 SEK


In everyday life, we need to selectively attend to task-relevant events while ignoring distracters. This top-down control is critical for selective attention, but its mechanisms are unclear. Working memory (WM) resources appear to be critical. In support, in individuals with low WM capacity or during a demanding WM task (high load), selective attention is typically impaired. However, recent behavioral studies made a counterintuitive observation: Selective attention during high WM load was actually improved in individuals with low WM capacity. Although counterintuitive, these findings are consistent with a modern theory of attention in which spatial attention has the shape of a Mexican hat. These previous behavioral effects are intriguing and important, but they do not demonstrate that distracter processing was actually affected, and do not permit causal inferences about the mechanism.

The proposed research will thoroughly examine the role of WM resources in selective attention. WM will be manipulated in terms of different tasks and individual differences in capacity. Combined with modern brain imaging methods (EEG, fMRI, MEG, and nTMS), four specific questions will be addressed: 1) effects of WM impairments on sensory processes of distracters and targets, 2) the mediating role of feedback processes in visual cortex, 3) the active ingredient in WM, and 4) the causal role of different parts of the frontal-parietal network.

Initially, several behavioral studies will be conducted (in the first 2 years) to identify a task paradigm (e.g., Flanker or Stroop) that has robust effects of WM manipulations (individual capacity and task load) on behavior. Because electroencephalography (EEG) is relatively cheap, these data will be concurrently collected in these initial studies to examine specific effects on distracter processing. Once a task paradigm is identified, a study with magnetoencephalography (MEG) will be conducted (in years 2 or 3) to study the mediating role of recurrent processing in visual cortex. Also, an fMRI study will be conducted to identify the neural correlates of selective attention with an emphasis on the frontal-parietal network. On the basis of these data, navigated transcranial magnetic stimulation (nTMS) will be used to target an identified brain area to examine its causal role in top-down control of selective attention. That is, effects of nTMS on different dependent variables (behavior, EEG, MEG) will be examined in different studies (in years 3 and 4). In addition, nTMS will already be used from the first year to target potential brain areas that have been suggested in the literature (however, the spatial resolution has been missing in previous studies with standard TMS). Also, another MEG study (year 1 or 2) will copy a published task design on recurrent processing and combine it with WM manipulations. For each study, within-subjects manipulations of WM load will require 20 subjects, and between-subjects manipulations of WM capacity will focus on extreme group analysis (i.e., low versus high). Thus, sample size will be around 40 per study. Although I have extensive experience with EEG and fMRI, I will need assistance with MEG and nTMS. To conduct MEG, I will have access to and help from NatMEG (the national facility for MEG). To conduct nTMS, I will collaborate with Dr. Mominul Islam at Karolinska Hospital, where the brand new nTMS system was just installed this January.

In sum, the proposed research examines the causal role of WM in selective attention. Specifically, this research will address effects of WM on sensory processing, the potential mediator, the active ingredient in the top-down signal, and the causal role of specific parts of the frontal-parietal network. These results will advance our basic understanding of the top-down control in selective attention and thus, provide important groundwork for research on difficulties in top-down control.

External Partners

Last updated on 2017-07-04 at 09:23