Parchi zoologici ed acquari di tutto il mondo sono alla continua ricerca di strumenti per misurare in modo affidabile il benessere degli animali. L'impresa risulta essere complessa per via della difficile interpretazione di variazioni fisiologiche, psicologiche e comportamentali. Oggigiorno il benessere viene definito come l'equilibrio tra esperienze positive e negative; si cerca, quindi, di soddisfare bisogni specie-specifici e preferenze individuali. Recenti studi hanno evidenziato la preferenza degli animali nei confronti di ambienti prevedibili, grazie anche alla possibilità di esprimere i cosiddetti Comportamenti Anticipatori (CA): la sola espressione di questi pattern prima di eventi positivi sembra essere di per sé premiante. Attraverso questo studio si vuole trovare una relazione tra CA e cinque diversi contesti in un gruppo di tursiopi in ambiente controllato, con l'obiettivo di confrontare il diverso stato motivazionale degli animali. La raccolta dati è stata effettuata da novembre 2017 ad aprile 2018 nel Parco Zoologico dell'Attica, Grecia. Il modello di studio è rappresentato da un gruppo di sette tursiopi. La raccolta è stata divisa in due periodi: nel primo periodo è stata video-registrata l'attività anticipatoria prima delle presentazioni pubbliche e si sono condizionati quattro eventi attraverso l'introduzione di un segnale acustico unico per ogni contesto. Gli eventi condizionati sono stati: AA strutturale, AA strutturale con trainers, AA strutturale con persone non famigliari ed una condizione di controllo in cui non accadeva nulla. La durata dell'intervallo tra segnale acustico ed i contesti è stata portata a 4 minuti. Nel secondo periodo sono stati video-registrati i comportamenti anticipatori prima dei quattro contesti precedentemente condizionati. I video sono stati analizzati con il software BORIS. L'etogramma prende in considerazione i seguenti comportamenti: porpoising (P), pivot dive, looking (L), head-up swim (H), spy hoping e logging. Per valutare se i diversi contesti influissero quantità e tipo di comportamento anticipatorio, è stato creato un Generalized Linear Mixed Models (GLMM) per ogni pattern comportamentale, con un consecutivo post hoc pairwise per valutare in che modo la manifestazione di un dato comportamento differisse tra i contesti specifici. Tutte le analisi statistiche sono state effettuate con RStudio in ambiente R. I risultati mostrano una variazione significativa di tre dei comportamenti considerati in funzione dei contesti. In particolare, la frequenza dell'evento P (GLMM full vs null: χ2 = 45,559, df = 4, p < 0,001) risulta significativamente più alta prima delle presentazioni rispetto agli altri contesti. La durata degli stati L (χ2 = 23,58, df = 4, p < 0,001) ed H (χ2 = 52,28, df = 4, p < 0,001) risulta maggiore prima delle tre sessioni di AA rispetto ai contesti pre-presentazioni e di controllo. I nostri risultati dimostrano che i comportamenti anticipatori sono utili alla valutazione dello stato motivazionale dei tursiopi in ambiente controllato. Ad esempio, la maggior frequenza di P prima delle presentazioni suggerisce un livello di arousal elevato, forse dovuto all'imminente assunzione di cibo od all'inizio di una attività fisica più intensa. Analogamente, comportamenti anticipatori poco energetici svolti sulla superficie (L - H) indicano un maggior livello di sorveglianza dovuto all'introduzione di oggetti e/o soggetti non sempre disponibili nell'ambiente. Nel complesso, il seguente lavoro rappresenta un primo e significativo passo verso l'utilizzo dei comportamenti anticipatori come strumento per la gestione e la valutazione del benessere dei tursiopi in ambiente controllato.
Zoos and Aquaria worldwide are looking for reliable tools to measure animal welfare. The matter results highly complex as physiological, psychological and behavioural variations are difficult to interpret. Nowadays welfare is defined as the balance between positive and negative experiences. Recent evidences show that animals prefer predictable environments due to the possibility to display preparatory responses, also called Anticipatory Behaviours (ABs): the mere expression of these behavioural patterns before positive events seem to be rewarding itself. This study tries to link ABs with five different contexts in a group of seven bottlenose dolphins under human care, with the goal to compare animals' emotional states across different conditions. The data collection, which has occurred in Attica Zoological Park, Greece, from November 2017 to April 2018, was divided into two periods. The first period consisted in the recording of food-anticipatory activity and the conditioning of four events through the introduction of a specific condition-dependent acoustic signal. The conditioned contexts were structural EE, social and structural EE with trainer, social and structural EE with unfamiliar humans and an event where nothing is going to happen. Once ensured that the dolphins were able to associate the presence of the signal with the start of the relative context, the duration of the interval was increased until four minutes. The second period consisted in the recording of ABs prior to the four conditioned contexts. Every animal has been recorded for 19 observations per condition, which means a total of 2716 minutes of video analysed with BORIS. The ethogram, adapted for this study, is inspired to the literature and takes into consideration the following behaviours: porpoising (p), pivot dive, looking (l), head-up swim (h), spy hoping and logging. To understand if the contexts influenced the quantity and type of AB, it has been created a Generalized Linear Mixed Models (GLMM) for each behaviour, which represented the fixed factor. The observations, with the subjects considered as nested factors, were included as random factors. Finally, to evaluate how different conditions influenced the display of a certain behaviour, it has been conducted a post hoc pairwise test on the significant results. All the statistical analyses were conducted with RStudio in R environment. The results showed a considerable context-dependent variation of three behavioural patterns. In particular, the frequency of the event p (GLMM full vs null: χ2 = 45.559, df = 4, P < 0.001) resulted higher prior to public presentation than to other contexts. The duration of the states l (χ2 = 23.58, df = 4, P < 0.001) and h (χ2 = 52.28, df = 4, P < 0.001) resulted higher prior to the three EE contexts compared to public presentation and control condition. The results indicate that ABs are useful in assessing emotional states in captive bottlenose dolphins. The higher occurrence of p prior to public presentations suggests a higher level of arousal, perhaps due to imminent food intake or in preparation of high intensity physical activity. Similarly, low energy surface ABs (l - h) indicate a higher level of vigilance owed to the introduction of devices and/or subjects not always available in the environment. Overall, this study represent an initial and significant step toward the use of anticipatory behaviours as a tool to manage and assess welfare in bottlenose dolphins.
Anticipatory Behaviours Across Different Contexts: Measuring Motivation in Tursiops truncatus
ZIMMITTI, MARTINA
2017/2018
Abstract
Zoos and Aquaria worldwide are looking for reliable tools to measure animal welfare. The matter results highly complex as physiological, psychological and behavioural variations are difficult to interpret. Nowadays welfare is defined as the balance between positive and negative experiences. Recent evidences show that animals prefer predictable environments due to the possibility to display preparatory responses, also called Anticipatory Behaviours (ABs): the mere expression of these behavioural patterns before positive events seem to be rewarding itself. This study tries to link ABs with five different contexts in a group of seven bottlenose dolphins under human care, with the goal to compare animals' emotional states across different conditions. The data collection, which has occurred in Attica Zoological Park, Greece, from November 2017 to April 2018, was divided into two periods. The first period consisted in the recording of food-anticipatory activity and the conditioning of four events through the introduction of a specific condition-dependent acoustic signal. The conditioned contexts were structural EE, social and structural EE with trainer, social and structural EE with unfamiliar humans and an event where nothing is going to happen. Once ensured that the dolphins were able to associate the presence of the signal with the start of the relative context, the duration of the interval was increased until four minutes. The second period consisted in the recording of ABs prior to the four conditioned contexts. Every animal has been recorded for 19 observations per condition, which means a total of 2716 minutes of video analysed with BORIS. The ethogram, adapted for this study, is inspired to the literature and takes into consideration the following behaviours: porpoising (p), pivot dive, looking (l), head-up swim (h), spy hoping and logging. To understand if the contexts influenced the quantity and type of AB, it has been created a Generalized Linear Mixed Models (GLMM) for each behaviour, which represented the fixed factor. The observations, with the subjects considered as nested factors, were included as random factors. Finally, to evaluate how different conditions influenced the display of a certain behaviour, it has been conducted a post hoc pairwise test on the significant results. All the statistical analyses were conducted with RStudio in R environment. The results showed a considerable context-dependent variation of three behavioural patterns. In particular, the frequency of the event p (GLMM full vs null: χ2 = 45.559, df = 4, P < 0.001) resulted higher prior to public presentation than to other contexts. The duration of the states l (χ2 = 23.58, df = 4, P < 0.001) and h (χ2 = 52.28, df = 4, P < 0.001) resulted higher prior to the three EE contexts compared to public presentation and control condition. The results indicate that ABs are useful in assessing emotional states in captive bottlenose dolphins. The higher occurrence of p prior to public presentations suggests a higher level of arousal, perhaps due to imminent food intake or in preparation of high intensity physical activity. Similarly, low energy surface ABs (l - h) indicate a higher level of vigilance owed to the introduction of devices and/or subjects not always available in the environment. Overall, this study represent an initial and significant step toward the use of anticipatory behaviours as a tool to manage and assess welfare in bottlenose dolphins.File | Dimensione | Formato | |
---|---|---|---|
809583_martina_zimmitti_eng.pdf
non disponibili
Tipologia:
Altro materiale allegato
Dimensione
3.44 MB
Formato
Adobe PDF
|
3.44 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14240/100418