Background: Attentional biases toward body- and food-related stimuli are key mechanisms in maintaining Anorexia Nervosa (AN). However, traditional reaction-time indices from dot-probe tasks show low reliability and limited ability to distinguish between cognitive, decisional, and motor components of attentional processing. Objective: This study applies the Drift-Diffusion Model (DDM) to obtain a more accurate measure of attentional mechanisms, by disentangling perceptual, decisional, and executive components of response. Methods: Behavioural data from dot-probe tasks with food stimuli (high vs. low calorie) and body stimuli (underweight, normal weight, overweight) were analyzed in AN patients and healthy controls. The DDM extracted specific parameters: non-decision time (t₀), drift rate (v), decision threshold (a), and execution speed (d). Results: The attentional bias index (t₀ incongruent - t₀ congruent) showed no significant group differences, indicating the absence of an early automatic bias. Nonetheless, stimulus-specific patterns emerged: in food tasks, high-calorie stimuli increased non-decision times (perceptual interference), while incongruent trials showed higher decision efficiency (increased v) and shorter response times, consistent with strategic avoidance. In body tasks, overweight bodies reduced decisional efficiency across groups, while AN patients displayed longer overall response times for underweight bodies, suggesting more prolonged affective processing toward stimuli congruent with the disorder’s body ideal. The model demonstrated excellent fit across all conditions. Conclusions: The DDM proved a promising tool for studying cognitive mechanisms in AN, revealing strategic avoidance and top-down hypercontrol rather than automatic vigilance. These findings provide a refined understanding of attentional and decisional dynamics toward clinically relevant stimuli, with implications for theoretical reconceptualization and targeted therapeutic interventions.
I bias attentivi verso stimoli corporei e alimentari rappresentano un meccanismo cruciale nel mantenimento dell’anoressia nervosa (AN), una delle malattie psichiatriche con la più alta mortalità. Tuttavia, la letteratura recente mostra come gli indici tradizionali basati sui tempi di reazione nei compiti dot-probe presentino una scarsa affidabilità e una limitata capacità di isolare i processi attentivi sottostanti. Il presente studio adotta un approccio di modellizzazione computazionale per analizzare i dati comportamentali provenienti da diversi compiti dot-probe somministrati a un campione clinico di pazienti con AN. Mediante il Drift-Diffusion Model (DDM) è stato ricavato un indice di bias attentivo (t₀_incongruente - t₀_congruente), che rappresenta il tempo necessario per orientare l’attenzione verso o lontano dagli stimoli salienti (immagini di cibo o di corpi). I modelli DDM hanno mostrato un buon fitting e un’affidabilità superiore rispetto agli indici tradizionali, confermando e ampliando i risultati precedentemente riportati in letteratura. Questi dati evidenziano il valore della psichiatria computazionale nello studio dell’anoressia nervosa, fornendo una misura più precisa e interpretabile dei meccanismi attentivi e offrendo nuove prospettive sulla relazione tra processi cognitivi, rappresentazione corporea e sintomi alimentari del disturbo.
Pattern attentivi nell'anoressia nervosa: un approccio computazionale
GIOVANNINI, SERENA
2023/2024
Abstract
Background: Attentional biases toward body- and food-related stimuli are key mechanisms in maintaining Anorexia Nervosa (AN). However, traditional reaction-time indices from dot-probe tasks show low reliability and limited ability to distinguish between cognitive, decisional, and motor components of attentional processing. Objective: This study applies the Drift-Diffusion Model (DDM) to obtain a more accurate measure of attentional mechanisms, by disentangling perceptual, decisional, and executive components of response. Methods: Behavioural data from dot-probe tasks with food stimuli (high vs. low calorie) and body stimuli (underweight, normal weight, overweight) were analyzed in AN patients and healthy controls. The DDM extracted specific parameters: non-decision time (t₀), drift rate (v), decision threshold (a), and execution speed (d). Results: The attentional bias index (t₀ incongruent - t₀ congruent) showed no significant group differences, indicating the absence of an early automatic bias. Nonetheless, stimulus-specific patterns emerged: in food tasks, high-calorie stimuli increased non-decision times (perceptual interference), while incongruent trials showed higher decision efficiency (increased v) and shorter response times, consistent with strategic avoidance. In body tasks, overweight bodies reduced decisional efficiency across groups, while AN patients displayed longer overall response times for underweight bodies, suggesting more prolonged affective processing toward stimuli congruent with the disorder’s body ideal. The model demonstrated excellent fit across all conditions. Conclusions: The DDM proved a promising tool for studying cognitive mechanisms in AN, revealing strategic avoidance and top-down hypercontrol rather than automatic vigilance. These findings provide a refined understanding of attentional and decisional dynamics toward clinically relevant stimuli, with implications for theoretical reconceptualization and targeted therapeutic interventions.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI_SG-2.pdf
Accesso riservato
Dimensione
796.69 kB
Formato
Adobe PDF
|
796.69 kB | Adobe PDF |
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/20.500.12608/97024