Interacting rhythms enhance sensitivity of target detection in a fronto-parietal computational model of visual attention
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This paper is of interest to neuroscientists studying the interaction between working memory, decision making, cell types, and neural oscillations. It introduces a detailed model of different brain areas which interact giving rise to the complex pattern of oscillations that are observed during a visual attention task. Additionally, the model reproduces the phase-dependent behavioral performance observed experimentally during such a task. This provides a new level of precision in our understanding of how rhythmic attention works in the brain.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Neuroscience (eLife)
Abstract
Even during sustained attention, enhanced processing of attended stimuli waxes and wanes rhythmically, with periods of enhanced and relatively diminished visual processing (and subsequent target detection) alternating at 4 or 8 Hz in a sustained visual attention task. These alternating attentional states occur alongside alternating dynamical states, in which lateral intraparietal cortex (LIP), the frontal eye field (FEF), and the mediodorsal pulvinar (mdPul) exhibit different activity and functional connectivity at α, β, and γ frequencies—rhythms associated with visual processing, working memory, and motor suppression. To assess whether and how these multiple interacting rhythms contribute to periodicity in attention, we propose a detailed computational model of FEF and LIP. When driven by θ-rhythmic inputs simulating experimentally-observed mdPul activity, this model reproduced the rhythmic dynamics and behavioral consequences of observed attentional states, revealing that the frequencies and mechanisms of the observed rhythms allow for peak sensitivity in visual target detection while maintaining functional flexibility.
Article activity feed
-
-
-
-
Evaluation Summary:
This paper is of interest to neuroscientists studying the interaction between working memory, decision making, cell types, and neural oscillations. It introduces a detailed model of different brain areas which interact giving rise to the complex pattern of oscillations that are observed during a visual attention task. Additionally, the model reproduces the phase-dependent behavioral performance observed experimentally during such a task. This provides a new level of precision in our understanding of how rhythmic attention works in the brain.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
-
Reviewer #1 (Public Review):
This work introduces a new detailed computational model that can reproduce the patterns of neural oscillations in the frontal eye field (FEF) and in the lateral intraparietal cortex (LIP) as a result of their mutual interaction and the inputs from other brain areas. In particular, the model matches the experimentally recorded periodic change of neural activity frequency bands during the delay interval. Finally, the model can reproduce the theta-phase dependent behavioral performance observed experimentally in a previous study by the authors (Fiebelkorn et al., 2019).
.
Strengths:
The model captures key empirical observations while incorporating several realistic biological features. In particular, the model investigates how oscillatory dynamics emerge from the interaction of different cortical cell types, …Reviewer #1 (Public Review):
This work introduces a new detailed computational model that can reproduce the patterns of neural oscillations in the frontal eye field (FEF) and in the lateral intraparietal cortex (LIP) as a result of their mutual interaction and the inputs from other brain areas. In particular, the model matches the experimentally recorded periodic change of neural activity frequency bands during the delay interval. Finally, the model can reproduce the theta-phase dependent behavioral performance observed experimentally in a previous study by the authors (Fiebelkorn et al., 2019).
.
Strengths:
The model captures key empirical observations while incorporating several realistic biological features. In particular, the model investigates how oscillatory dynamics emerge from the interaction of different cortical cell types, which is an important question in modern neuroscience.Despite the complexity of the model, this work provides a mechanistic explanation of some of the observed phenomena by taking apart different modular components of the full network.
Weaknesses:
It is not always clear how some of the model architecture and parameters were selected. Therefore it is at times difficult to distinguish experimentally based assumptions from model predictions.This study includes analyses of how key features depend on model parameters. However, there are only a limited number of such analyses, decreasing the generalizability of the observed phenomena to different model structures.
-
Reviewer #2 (Public Review):
This paper presents a biologically detailed model of LIP and FEF using single/multiple compartment neuron models. The authors showed that the model is able to produce the oscillation activities observed experimentally. Specifically, FEF and LIP modules are capable of producing β_2 and γ band synchronous firings, respectively when they are driven by the 13 Hz mdPul stimulus (aiming to mimic the good θ phase), while only LIP produces rhythm at the β_1 band without the mdPul stimulation (mimicking the bad θ phase).
Then, by incorporating an attention task used in biological experiments, the authors found that the model is able to reproduce the observed theta-rhythm shift of hit rates (Figure 7C) when receiving theta-rhythmic inputs from mdPul and V4 during the cue-target interval. Moreover, increasing the LIP → …
Reviewer #2 (Public Review):
This paper presents a biologically detailed model of LIP and FEF using single/multiple compartment neuron models. The authors showed that the model is able to produce the oscillation activities observed experimentally. Specifically, FEF and LIP modules are capable of producing β_2 and γ band synchronous firings, respectively when they are driven by the 13 Hz mdPul stimulus (aiming to mimic the good θ phase), while only LIP produces rhythm at the β_1 band without the mdPul stimulation (mimicking the bad θ phase).
Then, by incorporating an attention task used in biological experiments, the authors found that the model is able to reproduce the observed theta-rhythm shift of hit rates (Figure 7C) when receiving theta-rhythmic inputs from mdPul and V4 during the cue-target interval. Moreover, increasing the LIP → FEF visual module connection strength leads to LIP β_1 synchronous firing in the poor θ phase is able to excite FEF visual neurons to make a target detection, thus producing 8 Hz hit rate rhythm (Figure 8B). In addition, the authors disturbed oscillation activities in FEF and LIP modules computationally, and found that such manipulation alters the model performance on target detection, demonstrating that the rhythmic activities produced in FEF and LIP have their functional significance.
Finally, the authors showed that their model is robust enough to parameter changes for target detection, while is also flexible enough to produce different oscillations at other frequencies.The main strength of this paper is providing a biologically detailed model of LIP and FEF for generating the experimentally observed rhythmic activities, and demonstrating their potential computational roles. It is a valuable contribution to the field. The strength but also the limitation of this work is that the model is so complex, which makes readers hard to catch the mechanistic insight from the model simulation. Also, the rational of choosing some key parameters is very unclear.
-
-