FV 2 Mapping encoding and flow of task-relevant information from human MEG recordings in a perceptual decision-making task
Standard
FV 2 Mapping encoding and flow of task-relevant information from human MEG recordings in a perceptual decision-making task. / Nili, Hamed; Celotto, Marco; Toso, Alessandro; Donner, Tobias; Panzeri, Stefano.
In: CLIN NEUROPHYSIOL, Vol. 148, 2023, p. e2.Research output: SCORING: Contribution to journal › Conference abstract in journal › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - FV 2 Mapping encoding and flow of task-relevant information from human MEG recordings in a perceptual decision-making task
AU - Nili, Hamed
AU - Celotto, Marco
AU - Toso, Alessandro
AU - Donner, Tobias
AU - Panzeri, Stefano
PY - 2023
Y1 - 2023
N2 - Background: What are the brain computations used to take appropriatedecisions based on the available sensory evidence? Recordingsof cortical population activity of subjects performing perceptual discriminationtasks have begun to reveal the distribution across brainareas and frequency bands of neural signals encoding the sensorystimulus or the subject’s decision (Wilming et al., Nat. Commun.,2020), as well as mapping the transmission of activity betweenselected pairs of visual regions (Bosman et al., Neuron, 2012; Ferroet al., PNAS, 2021). Despite this progress, a comprehensive and quantitativelarge-scale mapping of both information encoding and transmissionduring perceptual decisions is still lacking.Objective: Our objective is to perform a complete identification ofwhen and how sensory information is encoded in the cortical hierarchyand in the frequency activity space, how it is transmitted toother brain areas, and how is used to guide behavioral choice.Methods: We analysed published whole-brain MEG recordings(Wilming et al., Nat. Commun., 2020) from human participants performinga visual contrast discrimination task entailing a sequence of10 discrete sample stimuli whose contrast had to be averaged andcompared to a reference. We used mutual information to quantifystimulus and choice information in brain activity across time, frequency,and 180 regions per hemisphere covering the entire cortex.We used Intersection Information (II, Pica et al., NIPS, 2017) to quantifyhow much of the sensory information encoded in neural activityis used to inform choices. Moreover, we measured the amount ofinformation about stimulus or choice transmitted between corticalareas using our newly developed Feature-specific InformationTransfer (FIT) measure.Results: We identified frequency bands that carried stimulus, choice,and intersection information across cortex. Stimulus informationused to inform choices was initially carried in the gamma-band,[40, 80) Hz, in visual, parietal and posterior cingulate regions.Choice signals later developed in the alpha-/beta-bands, [8, 40) Hz,in downstream regions of premotor and somato-motor cortex. Theregion-specific time-frequency patterns of information accuratelypredicted the region’s placement in the anatomical hierarchy andhad distinguishable profiles for different anatomical groups.Stimulus-specific and choice-specific information were broadlytransmitted across many areas, with transmission of stimulus informationoccurring predominantly in the feedforward direction in thegamma-band, and transmission of choice information predominantlyin the feedback direction in the alpha-/beta-bands.Conclusions: Using information theory, we mapped the encodingand transmission of task-relevant variables across frequency andcortical locations in the human brain during a perceptual discriminationtask. This revealed a transformation over time of stimuluinformation in the gamma band in visual, parietal and cingulateareas into choice signals in the beta and alpha band in downstreamareas, mediated by distributed patterns of feedforward gamma stimulusinformation and feedback alpha-beta choice informationtransmission.
AB - Background: What are the brain computations used to take appropriatedecisions based on the available sensory evidence? Recordingsof cortical population activity of subjects performing perceptual discriminationtasks have begun to reveal the distribution across brainareas and frequency bands of neural signals encoding the sensorystimulus or the subject’s decision (Wilming et al., Nat. Commun.,2020), as well as mapping the transmission of activity betweenselected pairs of visual regions (Bosman et al., Neuron, 2012; Ferroet al., PNAS, 2021). Despite this progress, a comprehensive and quantitativelarge-scale mapping of both information encoding and transmissionduring perceptual decisions is still lacking.Objective: Our objective is to perform a complete identification ofwhen and how sensory information is encoded in the cortical hierarchyand in the frequency activity space, how it is transmitted toother brain areas, and how is used to guide behavioral choice.Methods: We analysed published whole-brain MEG recordings(Wilming et al., Nat. Commun., 2020) from human participants performinga visual contrast discrimination task entailing a sequence of10 discrete sample stimuli whose contrast had to be averaged andcompared to a reference. We used mutual information to quantifystimulus and choice information in brain activity across time, frequency,and 180 regions per hemisphere covering the entire cortex.We used Intersection Information (II, Pica et al., NIPS, 2017) to quantifyhow much of the sensory information encoded in neural activityis used to inform choices. Moreover, we measured the amount ofinformation about stimulus or choice transmitted between corticalareas using our newly developed Feature-specific InformationTransfer (FIT) measure.Results: We identified frequency bands that carried stimulus, choice,and intersection information across cortex. Stimulus informationused to inform choices was initially carried in the gamma-band,[40, 80) Hz, in visual, parietal and posterior cingulate regions.Choice signals later developed in the alpha-/beta-bands, [8, 40) Hz,in downstream regions of premotor and somato-motor cortex. Theregion-specific time-frequency patterns of information accuratelypredicted the region’s placement in the anatomical hierarchy andhad distinguishable profiles for different anatomical groups.Stimulus-specific and choice-specific information were broadlytransmitted across many areas, with transmission of stimulus informationoccurring predominantly in the feedforward direction in thegamma-band, and transmission of choice information predominantlyin the feedback direction in the alpha-/beta-bands.Conclusions: Using information theory, we mapped the encodingand transmission of task-relevant variables across frequency andcortical locations in the human brain during a perceptual discriminationtask. This revealed a transformation over time of stimuluinformation in the gamma band in visual, parietal and cingulateareas into choice signals in the beta and alpha band in downstreamareas, mediated by distributed patterns of feedforward gamma stimulusinformation and feedback alpha-beta choice informationtransmission.
U2 - 10.1016/j.clinph.2023.02.003
DO - 10.1016/j.clinph.2023.02.003
M3 - Conference abstract in journal
VL - 148
SP - e2
JO - CLIN NEUROPHYSIOL
JF - CLIN NEUROPHYSIOL
SN - 1388-2457
ER -