Attention refers to the brain mechanisms that select and prioritize relevant information from competing sensory signals, supporting adaptive and goal-directed behaviour. While substantial progress has been made in characterizing the functional architecture of attention, critical questions remain about how stable individual characteristics, dynamic contextual factors, and moment-to-moment neural processes interact to shape attentional function and conscious perception. The present thesis addresses these questions through three complementary empirical investigations that trace a path from bottom-up constitutional influences to top-down awareness, examining the neural architecture of attention at multiple levels of analysis. Study 1 investigated the overlap between attentional networks and temperament dimensions. Healthy adults completed the Attention Network Test while pupillary responses were recorded. Structure of Temperament Questionnaire (STQ-77) was administered. Individuals with lower ergonicity and cognitive integration abilities exhibited a more pronounced incongruency effect on manual reaction times. Lower scores in temperamental traits belonging to the orientation dimension resulted in reduced accuracy in incongruent trials. Within the cue-target interval, individuals with a high level of neuroticism exhibited sustained pupil dilation. During the same interval, anticipatory dilation was found in individuals with lower probabilistic reasoning, reduced ability to engage in prolonged socio-verbal activities, and high inhibitory control. These findings support a temperament-based modulation of both behavioural and physiological correlates of attention, accounting for inter-individual differences. Moving on to Study 2, I examined how contextual expectations modulate neural activity underlying spatial attention using intracranial recordings from patients with drug-resistant epilepsy. Participants performed a Posner cueing task under Non-predictive, Predictive, and Counter-predictive conditions. Trajectory k-means clustering identified functionally distinct neural populations: an early cluster showing retinotopic sensitivity to contralateral stimuli and a later cluster exhibiting bilateral responses with context-dependent modulation. The critical finding was enhanced neural responses to rare congruent trials in the Counter-predictive condition—a "surprise effect" reflecting the detection of events that violated learned contingencies—demonstrating that the neural dynamics of spatial attention are dynamically shaped by the statistical structure of cue-target relationships. Study 3 aimed to bridge intracerebral EEG events with corresponding surface recordings, differentiating between conscious and non-conscious processing. ERP analysis revealed a significant positive deflection for Seen compared to Unseen targets in the 350–500 ms post-target window at frontal sites, consistent with a P3b component associated with conscious report. Time-frequency analysis revealed spectral dynamics associated with conscious report, including pre-target beta/gamma power modulations at the frontal electrode and post-target increased oscillatory activity at the occipital site. Trajectory k-means clustering of intracerebral data identified two key patterns of post-target activity: a Visual cluster exhibiting early, transient responses and an Accumulation cluster demonstrating gradual activity buildup. Ridge regression analysis revealed that, compared to the Visual cluster, the Accumulation cluster contributed more to the prediction of report-related ERPs at the scalp level. These findings highlight the greater contribution of late integrative mechanisms, compared to early sensory processes, in the conscious experience of behaviourally relevant targets. Together, they reveal attention as a multi-level architecture in which constitutional predispositions, contextual expectations, and late integrative neural processes jointly determine both attentional efficiency and conscious access.

From bottom-up traits to top-down awareness: the neural architecture of attention / Lozito, Silvana. - (2026 May 18).

From bottom-up traits to top-down awareness: the neural architecture of attention

LOZITO, SILVANA
18/05/2026

Abstract

Attention refers to the brain mechanisms that select and prioritize relevant information from competing sensory signals, supporting adaptive and goal-directed behaviour. While substantial progress has been made in characterizing the functional architecture of attention, critical questions remain about how stable individual characteristics, dynamic contextual factors, and moment-to-moment neural processes interact to shape attentional function and conscious perception. The present thesis addresses these questions through three complementary empirical investigations that trace a path from bottom-up constitutional influences to top-down awareness, examining the neural architecture of attention at multiple levels of analysis. Study 1 investigated the overlap between attentional networks and temperament dimensions. Healthy adults completed the Attention Network Test while pupillary responses were recorded. Structure of Temperament Questionnaire (STQ-77) was administered. Individuals with lower ergonicity and cognitive integration abilities exhibited a more pronounced incongruency effect on manual reaction times. Lower scores in temperamental traits belonging to the orientation dimension resulted in reduced accuracy in incongruent trials. Within the cue-target interval, individuals with a high level of neuroticism exhibited sustained pupil dilation. During the same interval, anticipatory dilation was found in individuals with lower probabilistic reasoning, reduced ability to engage in prolonged socio-verbal activities, and high inhibitory control. These findings support a temperament-based modulation of both behavioural and physiological correlates of attention, accounting for inter-individual differences. Moving on to Study 2, I examined how contextual expectations modulate neural activity underlying spatial attention using intracranial recordings from patients with drug-resistant epilepsy. Participants performed a Posner cueing task under Non-predictive, Predictive, and Counter-predictive conditions. Trajectory k-means clustering identified functionally distinct neural populations: an early cluster showing retinotopic sensitivity to contralateral stimuli and a later cluster exhibiting bilateral responses with context-dependent modulation. The critical finding was enhanced neural responses to rare congruent trials in the Counter-predictive condition—a "surprise effect" reflecting the detection of events that violated learned contingencies—demonstrating that the neural dynamics of spatial attention are dynamically shaped by the statistical structure of cue-target relationships. Study 3 aimed to bridge intracerebral EEG events with corresponding surface recordings, differentiating between conscious and non-conscious processing. ERP analysis revealed a significant positive deflection for Seen compared to Unseen targets in the 350–500 ms post-target window at frontal sites, consistent with a P3b component associated with conscious report. Time-frequency analysis revealed spectral dynamics associated with conscious report, including pre-target beta/gamma power modulations at the frontal electrode and post-target increased oscillatory activity at the occipital site. Trajectory k-means clustering of intracerebral data identified two key patterns of post-target activity: a Visual cluster exhibiting early, transient responses and an Accumulation cluster demonstrating gradual activity buildup. Ridge regression analysis revealed that, compared to the Visual cluster, the Accumulation cluster contributed more to the prediction of report-related ERPs at the scalp level. These findings highlight the greater contribution of late integrative mechanisms, compared to early sensory processes, in the conscious experience of behaviourally relevant targets. Together, they reveal attention as a multi-level architecture in which constitutional predispositions, contextual expectations, and late integrative neural processes jointly determine both attentional efficiency and conscious access.
18-mag-2026
File allegati a questo prodotto
File Dimensione Formato  
Tesi_dottorato_Lozito.pdf

accesso aperto

Note: tesi completa
Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 39.72 MB
Formato Adobe PDF
39.72 MB Adobe PDF

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1768156
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact