HOW DO WE BECOME AWARE OF WHAT (WE THINK) WE SEE?
Oscillatory mechanisms of conscious perception
Jelena Trajkovic1, Francesco Di Gregorio2, Paolo Di Luzio1, Eleonora Mercantoni1, Alessio Avenanti1, Gregor Thut3, Vincenzo Romei1,4
1 Centro studi e ricerche in Neuroscienze Cognitive, Dipartimento di Psicologia, Alma Mater Studiorum – Università di Bologna, Campus di Cesena, 47521 Cesena, Italy;
2 Azienda Unità Sanitaria Locale, UOC Medicina riabilitativa e neuroriabilitazione, 40139, Bologna, Italy;
3Institute of Neuroscience & Psychology, University of Glasgow, UK
4 IRCCS Fondazione Santa Lucia, 00179 Roma, Italy.
Alpha-oscillations have an active role in sensory processing and conscious perception. Specifically, recent reports suggest that pre-stimulus alpha-amplitude reflects perceptual bias, not accuracy, during visual decision-making. What are then oscillatory correlates of perceptual accuracy? Individual alpha-frequency (IAF) seems to have a relevant role in temporal sampling, such that faster alpha-oscillations lead to higher accumulation of sensory evidence over time, resulting in more accurate perceptual experience. Importantly, we hypothesize here that this higher temporal resolution for visual sampling can successfully translate into higher spatial mapping, by allocating more resources in space. In the present work, we used a visual detection task with spatially lateralized stimuli, firstly to directly test the hypothesis that IAF and alpha-amplitude selectively account for objective accuracy and subjective confidence, respectively. Moreover, in a second experiment, we independently targeted pre-stimulus IAF and amplitude by rhythmic Transcranial Magnetic Stimulation protocols, thus testing how changes in these two parameters of alpha lead to distinct behavioral changes. We found that accuracy is best predicted by modulations of the pre-stimulus IAF. Specifically, induced faster vs. slower alpha-TMS oscillations directly lead to higher vs. lower task accuracy, not confidence. Finally, we causally substantiated the emerging view that alpha-amplitude modulations selectively reflect the degree of subjective confidence, with IAF-tuned TMS-pulses aimed at enhancing alpha-amplitude leading to lower confidence ratings. Here, we offer a first causal evidence of a functional dissociation between distinct parameters of alpha-activity in shaping conscious perception: while IAF accounts for the accuracy of our perceptual experience, alpha-amplitude shapes its’ subjective interpretation.
Driving associative plasticity in reentrant visual networks dissociate subjective vs. objective components of perceptual decision-making
Paolo Di Luzio1, Luca Tarasi1, Juha Silvanto2, Alessio Avenanti1, Vincenzo Romei1
1. University of Bologna, Italy; 2. University of Surrey, UK
Perceptual decisions have different components, such as sensitivity to the stimulus and response confidence. Here we show that these components rely on functionally distinct networks. Reentrant projections from motion visual area (V5) to primary visual area (V1) have been shown to represent the neural substrate accounting for conscious experience of motion visual stimuli both in animals and humans. The macaque lateral intraparietal area (LIP), homologue of the human intraparietal sulcus (IPS) has been shown to play a crucial role in forming confidence. Accordingly, transiently enhancing feedback connectivity from V5 to V1 by means of cortico-cortical paired associative stimulation (ccPAS) – a neurostimulation protocol inducing transient modulation of brain connectivity according to spike‐timing dependent plasticity (STDP) mechanisms – can lead to improved motion discrimination performance. Here we tested for the first time a functional dissociation between V5-to-V1 and IPS-to-V1 back-projections by probing both individual accuracy and confidence in a motion discrimination task performance following V5-to-V1 or IPS-to-V1 ccPAS. We found that perceptual accuracy selectively increased following V5-to-V1 (but not IPS-toV1) ccPAS, leading to a higher number of correct responses, in line with previous findings. Perceptual confidence, on the other hand, significantly, and functionally increased for correct trials only following IPS-to-V1 ccPAS. Instead a general and unspecific increase in confidence was found, irrespective of response correctness, following V5-to-V1 ccPAS. These findings provide a first direct demonstration of a double dissociation of functional networks orchestrating perceptual decision-making in humans, namely V5-to-V1, accounting for conscious visual motion discrimination sensitivity, and IPS-to-V1, accounting for accurate confidence judgments.
Boosting working memory capacity by strengthening functional oscillatory pathways
Joaquín Macedo-Pascual1,2, Paolo Di Luzio1, Claudia Poch3, Vincenzo Romei1
1. Università di Bologna, Bologna, Italy; 2. Universidad Complutense de Madrid, Madrid, Spain; 3. Universidad Nebrija, Madrid, Spain
Working Memory (WM) capacity is highly limited so top-down mechanisms play an essential role in enhancing relevant representations and inhibiting irrelevant information. Here, we aimed at targeting functionally dissociated oscillatory networks within the same neural substrate, i.e. the fronto-parietal network. Coherence in the alpha band (7-13 Hz) between fronto-parietal areas is associated with suppression of irrelevant information while coherence in the theta band (4-7 Hz) is related to retention of information. With the aim of enhancing connectivity between frontoparietal areas through enhanced coherence within either alpha or theta band, we applied a novel TMS protocol named cortico-cortical-paired-associative-stimulation (ccPAS) with inter-stimulus intervals (ISI) tuned on individual frequency-band peaks. A first TMS pulse was applied to a right frontal area followed at specific frequency-tuned ISI by a second pulse over a right parietal area. For the alpha-band group, we found an enhanced WM capacity for the right hemifield (ipsilaterally) attended items while for the theta-band group we found an enhanced WM capacity for the left hemifield (contralaterally) attended items.
Preliminary EEG analysis so far performed only in the alpha-band group showed an improvement in communication, both in terms of enhanced alpha (but not theta) coherence and reduced alpha (but not theta) power in the network. Our results stand in line with the expected outcome that stimulating with an ISI between frontal and parietal nodes corresponding to an individual alpha cycle would induce a selective inhibition of irrelevant items presented contralaterally, thus facilitating performance for the ipsilaterally attended items.
Stay tuned: driving plastic changes in visuo-motor chains through state-dependent cortico-cortical paired associative stimulation (ccPAS) over premotor-motor networks.
Turrini S.1, Fiori F. 1,2, Romei V.1,3, Avenanti A.1,3,4
- Centro studi e ricerche in Neuroscienze Cognitive, Dipartimento di Psicologia, Alma Mater Studiorum – Università di Bologna, 47521, Cesena, Italy.
- Dipartimento di Psicologia, Università di Milano-Bicocca, 20126, Milano, Italy.
- IRCCS Fondazione Santa Lucia, 00179, Rome, Italy.
- Centro de Investigación en Neuropsicología y Neurociencias Cognitivas, Universidad Católica Del Maule, 3460000, Talca, Chile.
Neural pathways from the ventral premotor cortex (PMv) to the primary motor cortex (M1) allow for sensorimotor transformations. Yet, whether brain stimulation over the PMv-to-M1 pathway can target and transiently modulate specific visuo-motor association remains to be determined. To address this issue, here, we used cortico-cortical paired associative stimulation (ccPAS), a TMS protocol able to enhance or hinder cortical connectivity between two stimulated sites through the mechanism of spike-timing-dependent-plasticity. We applied the protocol in state-dependent manner, to increase its functional specificity: ccPAS was administered over the PMv-M1 circuit during a simple action selection task engaging premotor-motor visuo-motor connections (index or little finger abduction in response to a target or a control visual stimulus). The ccPAS pulses were administered only during the execution of the movement associated with the presentation of the target stimulus, targeting one specific visuo-motor association. We assessed corticospinal excitability (CSE) before and after ccPAS, while participants remained at rest and passively viewed the same target/control stimuli shown during ccPAS. Participants underwent, on different days, PMv-to-M1 and M1-to-PMv ccPAS, to test the effects of supporting or contrasting the physiological activation of visuo-to-motor premotor-motor connections. We found that PMv-to-M1 ccPAS significantly enhanced CSE selectively for the visuomotor association targeted during ccPAS, leaving the control association unaltered; conversely, M1-to-PMv ccPAS effectively weakened the visuomotor association. These findings highlight a functional organization of the human PMv-to-M1 pathway in spatially overlapping but functionally specific visuo-motor circuits, and suggest that state-dependent ccPAS provides a tool for targeting such circuits and testing their malleability.