High-frequency transcranial random noise stimulation (hf-tRNS) has been proposed as a technique to enhance perceptual learning by modulating visual cortex excitability through mechanisms compatible with stochastic resonance. Whether the single-session advantage reported by Fertonani, Pirulli and Miniussi (2011) accumulates across distributed training protocols, a precondition for any clinical translation, remains an open question. This study replicates and extends Fertonani et al. (2011) by replacing the original single session with a ten-session protocol (one pre-test, eight training sessions, and one post-test) and adding an explicit transfer evaluation across four conditions orthogonal for orientation and retinal position (A-D). Twenty-four normally sighted adults, assigned to the two groups through systematic alternation by enrolment order (n = 12 per group), performed an orientation discrimination task (ODT) with online active hf-tRNS or sham stimulation over visual cortex. This preliminary analysis adopts a reduced plan with two analyses of sensitivity d′: a linear mixed-effects model for learning during training and a 2 × 2 × 4 mixed ANOVA for the pre-post change. The central finding is the absence of accumulation of the hf-tRNS advantage across the eight sessions (group × session in training: p = .562; group × time pre-post: p = .874). The only stimulation-related signal is a steeper tilt-related gradient of sensitivity in the active group (p = .007), which is, however, concentrated at the easy tilts (1.5°-2.2°) and already present at pre-test, where the groups differ on conditions B and C: it is therefore attributable to a baseline imbalance rather than to stimulation. Both groups show a robust gain on the trained condition (Δd′ = +0.234, p < .001), with no between-group difference and no differential transfer. Overall this preliminary analysis replicates perceptual learning but provides no evidence of a specific, cumulative hf-tRNS effect; the strongest claim the design supports is that reliable learning occurs on the trained condition, while conclusions about stimulation effects remain provisional and to be confirmed with extended analyses and larger samples.
La stimolazione elettrica transcranica a rumore casuale ad alta frequenza (hf-tRNS) è stata proposta come tecnica capace di potenziare l’apprendimento percettivo modulando l’eccitabilità della corteccia visiva attraverso meccanismi compatibili con la risonanza stocastica. Resta aperta la questione se il vantaggio osservato in singola sessione (Fertonani, Pirulli, & Miniussi, 2011) si accumuli lungo protocolli distribuiti, condizione necessaria per qualunque traduzione clinica. Lo studio replica ed estende il disegno di Fertonani et al. (2011) introducendo un protocollo su dieci sessioni (un pre-test, otto di addestramento e un post-test) e una valutazione del trasferimento su quattro condizioni ortogonali per orientamento e posizione retinica (A-D). Ventiquattro adulti con visione normale, assegnati in alternanza sistematica per ordine di arruolamento ai due gruppi (n = 12 per gruppo), hanno svolto un compito di discriminazione di orientamento con stimolazione hf-tRNS attiva o sham online sulla corteccia visiva. Questa analisi preliminare adotta un piano ridotto a due sole analisi della sensibilità d’: un modello lineare a effetti misti per l’apprendimento durante il training e un’ANOVA mista 2 × 2 × 4 per il cambiamento pre-post. Il risultato centrale è l’assenza di accumulo del vantaggio hf-tRNS lungo le otto sessioni (gruppo × sessione nell’addestramento: p = .562; gruppo × momento pre-post: p = .874). L’unico segnale legato alla stimolazione è un gradiente della sensibilità rispetto al tilt più ripido nel gruppo attivo (p = .007), concentrato però ai tilt facili (1.5°-2.2°) e già presente al pre-test, dove i gruppi differiscono in B e C: va quindi attribuito a uno squilibrio di baseline più che alla stimolazione. Entrambi i gruppi mostrano un guadagno robusto sulla condizione allenata (Δd’ = +0.234, p < .001), senza differenze tra gruppi e senza trasferimento differenziale. Nel complesso questa analisi preliminare replica l’apprendimento percettivo ma non documenta un effetto specifico e cumulativo della hf-tRNS; l’affermazione più solida che il disegno consente è l’esistenza di un apprendimento affidabile sulla condizione allenata, mentre le conclusioni sugli effetti della stimolazione restano provvisorie e da confermare con analisi estese e campioni più ampi.
Risonanza stocastica e plasticità sinaptica nell'apprendimento percettivo: effetti della stimolazione elettrica transcranica a rumore casuale su sessioni multiple
CANTAVENERA, DAVIDE
2025/2026
Abstract
High-frequency transcranial random noise stimulation (hf-tRNS) has been proposed as a technique to enhance perceptual learning by modulating visual cortex excitability through mechanisms compatible with stochastic resonance. Whether the single-session advantage reported by Fertonani, Pirulli and Miniussi (2011) accumulates across distributed training protocols, a precondition for any clinical translation, remains an open question. This study replicates and extends Fertonani et al. (2011) by replacing the original single session with a ten-session protocol (one pre-test, eight training sessions, and one post-test) and adding an explicit transfer evaluation across four conditions orthogonal for orientation and retinal position (A-D). Twenty-four normally sighted adults, assigned to the two groups through systematic alternation by enrolment order (n = 12 per group), performed an orientation discrimination task (ODT) with online active hf-tRNS or sham stimulation over visual cortex. This preliminary analysis adopts a reduced plan with two analyses of sensitivity d′: a linear mixed-effects model for learning during training and a 2 × 2 × 4 mixed ANOVA for the pre-post change. The central finding is the absence of accumulation of the hf-tRNS advantage across the eight sessions (group × session in training: p = .562; group × time pre-post: p = .874). The only stimulation-related signal is a steeper tilt-related gradient of sensitivity in the active group (p = .007), which is, however, concentrated at the easy tilts (1.5°-2.2°) and already present at pre-test, where the groups differ on conditions B and C: it is therefore attributable to a baseline imbalance rather than to stimulation. Both groups show a robust gain on the trained condition (Δd′ = +0.234, p < .001), with no between-group difference and no differential transfer. Overall this preliminary analysis replicates perceptual learning but provides no evidence of a specific, cumulative hf-tRNS effect; the strongest claim the design supports is that reliable learning occurs on the trained condition, while conclusions about stimulation effects remain provisional and to be confirmed with extended analyses and larger samples.| File | Dimensione | Formato | |
|---|---|---|---|
|
Cantavenera_Davide.pdf
accesso aperto
Dimensione
778.61 kB
Formato
Adobe PDF
|
778.61 kB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/110660