The present Master’s thesis was conducted with the aim of investigating, from a transdiagnostic perspective, the cognitive domain, particularly the styles of representation of reality, in individuals suffering from an eating disorder (ED). This investigation was conducted with a twofold aim: first, to update the existing literature with empirical data concerning the sphere of thought of individuals reporting eating-related symptomatology and than to support clinical work through the qualitative analysis of some cognitive patterns, in order to increase the understanding of the disorder itself and to guide psychotherapeutic work. As a theoretical basis, the areas concerning thoughts about food, weight and body shape, maladaptive perfectionism, emotion dysregulation, identity-related aspects, and body perception were considered. In this regard, the choice was made to focus on a level of cognition that is more easily accessible to the individual, namely that of automatic thoughts, formulating the items in a gender-neutral manner, so as to promote broader identification. The developed instrument was administered to a total of 236 participants, divided into two groups, namely the clinical group (N = 38) and the control sample (N = 198; consisting of individuals who did not report an ED). Following data collection, the following analyses were conducted: to assess the dimensionality of the scale and the fit of the model to the data, Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were performed, and the fit indices (CFI, TLI, RMSEA, and SRMR) were considered. Subsequently, the reliability (through Cronbach’s α and McDonald’s ω) and validity (criterion and construct validity, both convergent and divergent) of the developed scale were assessed. Four types of models (three correlated factors, bifactor, second-order, and unidimensional) were compared through factor analysis, and the unidimensional structure was identified as the optimal solution. The latter presents a solid factor structure, in which the items show generally strong degrees of association with the general factor, ranging from .250 to .923 (the majority show factor loadings above .70), and good psychometric properties, given that internal consistency is high (α = .95; ω = .97) and the scale proves to be valid. This statement derives from the fact that the data show that the scale is able to discriminate between clinical and non-clinical patients, given the presence of a statistically significant difference between the two groups (W = 2855.5, p < .001) in the degree of agreement with the automatic thoughts associated with eating-related symptomatology expressed by the participants. Furthermore, it meets the criteria for construct validity, both convergent, presenting strong-magnitude correlations with the EDE-Q questionnaire, and divergent, demonstrating very weak or null associations with the COPE NVI 25 and BFI 10 scales. To further investigate the cognitive level, the scores of the automatic thoughts questionnaire and the CDS scale were examined, highlighting moderate positive correlations between them; the clinical group also presents higher mean scores on the CDS than the healthy controls (statistically significant, p < .05) in nine of the ten dimensions of the scale. It was not possible to demonstrate the transdiagnostic nature of the validated scale, as the sample size, particularly that of the clinical group, was limited. In conclusion, it can be hypothesized that the instrument, following further analyses and an expansion of the sample, it could be used both in clinical and research settings, in order to further investigate the cognitive sphere of individuals reporting an eating disorder.
Il presente lavoro di tesi magistrale è stato realizzato con lo scopo di indagare, in ottica transdiagnostica, l'area cognitiva, in particolare gli stili di rappresentazione della realtà, di chi soffre di un disturbo del comportamento alimentare (DCA). Tale indagine è stata condotta con un duplice fine: aggiornare la letteratura esistente di dati empirici riguardo alla sfera del pensiero di chi riporta sintomatologia alimentare e supportare il lavoro clinico attraverso l'analisi qualitativa di alcuni schemi cognitivi, per aumentare la comprensione della patologia stessa e orientare il lavoro psicoterapico. Come base teorica, sono state prese in considerazione le aree concernenti i pensieri sul cibo, peso e forma corporea, il perfezionismo maladattivo, la disregolazione emotiva, gli aspetti identitari e la percezione corporea. A tal proposito, è stata compiuta la scelta di focalizzarsi su un livello di cognizione di più facile accesso alla persona, ossia quello dei pensieri automatici, formulando gli item con una modalità priva di connotazioni di genere, così da favorire un più ampio rispecchiamento. Lo strumento realizzato è stato somministrato ad un totale di 236 partecipanti, suddivisi in due gruppi, ovvero quello clinico (N= 38) e di controllo (N= 198; formato da individui non riportanti un DCA). A seguito della raccolta dati, sono state compiute le seguenti analisi: per la valutazione della dimensionalità della scala e l'adattamento del modello ai dati sono state effettuate l'Analisi Fattoriale Esplorativa (EFA) e Confermativa (CFA) e considerati gli indici di fit (CFI, TLI, RMSEA e SRMR). Successivamente sono state valutate l'attendibilità (tramite l’α di Cronbach e l'α McDonald) e la validità (di criterio e di costrutto, sia convergente che divergente) della scala realizzata. Sono state confrontate, attraverso l'analisi fattoriale, quattro tipologie di modelli (a tre fattori correlati, bifattoriale, di secondo ordine e monofattoriale) ed è stata individuata nella struttura monodimensionale la soluzione ottimale. Quest'ultima presenta una struttura fattoriale solida, in cui gli item presentano gradi di associazione tendenzialmente forti con il fattore generale, compresi tra .250 e .923 (la maggior parte degli mostra saturazioni superiori a .70), e buone proprietà psicometriche dato che la consistenza interna risulta elevata (α = .95; ω = .97) e risulta essere valido. Tale affermazione deriva dal fatto che i dati mostrano che la scala risulta in grado di discriminare tra pazienti clinici e non, vista la presenza di una differenza statisticamente significativa tra i due gruppi (W = 2855.5, p < .001) nel grado di accordo con i pensieri automatici associati alla sintomatologia alimentare espresso dai partecipanti. Inoltre, soddisfa i criteri per la validità di costrutto, sia convergente, presentando correlazioni di magnitudine forte con il questionario EDE-Q, che divergente, dimostrando associazioni molto deboli o nulle con le scale COPE NVI 25 e BFI 10. Per approfondire il livello cognitivo sono stati presi in esame i punteggi del questionario sui pensieri automatici e la scala CDS, evidenziando correlazioni positive di entità moderata tra di essi; il gruppo clinico inoltre presenta punteggi medi maggiori alla CDS rispetto ai controlli sani (statisticamente significativi p < .05) in nove delle dieci dimensioni della scala. Non è stato possibile dimostrare la natura transdiagnostica della scala validata, essendo la numerosità campionaria, in particolare quella del gruppo clinico, contenuta. Concludendo, è possibile ipotizzare che lo strumento, a seguito di ulteriori analisi e di un ampliamento del campione, potrebbe essere utilizzato sia in ambito clinico, sia di ricerca, al fine di approfondire la sfera cognitiva di chi riporta un disturbo del comportamento alimentare.
Disturbi del comportamento alimentare e pensieri automatici: un’indagine preliminare sulla validazione di uno strumento transdiagnostico.
SCHIVO, ANNA
2025/2026
Abstract
The present Master’s thesis was conducted with the aim of investigating, from a transdiagnostic perspective, the cognitive domain, particularly the styles of representation of reality, in individuals suffering from an eating disorder (ED). This investigation was conducted with a twofold aim: first, to update the existing literature with empirical data concerning the sphere of thought of individuals reporting eating-related symptomatology and than to support clinical work through the qualitative analysis of some cognitive patterns, in order to increase the understanding of the disorder itself and to guide psychotherapeutic work. As a theoretical basis, the areas concerning thoughts about food, weight and body shape, maladaptive perfectionism, emotion dysregulation, identity-related aspects, and body perception were considered. In this regard, the choice was made to focus on a level of cognition that is more easily accessible to the individual, namely that of automatic thoughts, formulating the items in a gender-neutral manner, so as to promote broader identification. The developed instrument was administered to a total of 236 participants, divided into two groups, namely the clinical group (N = 38) and the control sample (N = 198; consisting of individuals who did not report an ED). Following data collection, the following analyses were conducted: to assess the dimensionality of the scale and the fit of the model to the data, Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were performed, and the fit indices (CFI, TLI, RMSEA, and SRMR) were considered. Subsequently, the reliability (through Cronbach’s α and McDonald’s ω) and validity (criterion and construct validity, both convergent and divergent) of the developed scale were assessed. Four types of models (three correlated factors, bifactor, second-order, and unidimensional) were compared through factor analysis, and the unidimensional structure was identified as the optimal solution. The latter presents a solid factor structure, in which the items show generally strong degrees of association with the general factor, ranging from .250 to .923 (the majority show factor loadings above .70), and good psychometric properties, given that internal consistency is high (α = .95; ω = .97) and the scale proves to be valid. This statement derives from the fact that the data show that the scale is able to discriminate between clinical and non-clinical patients, given the presence of a statistically significant difference between the two groups (W = 2855.5, p < .001) in the degree of agreement with the automatic thoughts associated with eating-related symptomatology expressed by the participants. Furthermore, it meets the criteria for construct validity, both convergent, presenting strong-magnitude correlations with the EDE-Q questionnaire, and divergent, demonstrating very weak or null associations with the COPE NVI 25 and BFI 10 scales. To further investigate the cognitive level, the scores of the automatic thoughts questionnaire and the CDS scale were examined, highlighting moderate positive correlations between them; the clinical group also presents higher mean scores on the CDS than the healthy controls (statistically significant, p < .05) in nine of the ten dimensions of the scale. It was not possible to demonstrate the transdiagnostic nature of the validated scale, as the sample size, particularly that of the clinical group, was limited. In conclusion, it can be hypothesized that the instrument, following further analyses and an expansion of the sample, it could be used both in clinical and research settings, in order to further investigate the cognitive sphere of individuals reporting an eating disorder.| File | Dimensione | Formato | |
|---|---|---|---|
|
Schivo_Anna.pdf
accesso aperto
Dimensione
1.98 MB
Formato
Adobe PDF
|
1.98 MB | 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/20.500.12608/118837