Fear of novel and startling stimuli in domestic dogs

Dog aggression to humans is a serious problem and it appears that fear-induced aggression in dogs is often implicated. Furthermore, it has been suggested that novel and startling stimuli may trigger fear-induced aggression. The aim of the present study was to study some possible behavioural and physiological measures of fear in the dog Canis familiaris in tests involving novel and startling stimuli. Dogs (n = 108) of various breeds, sex and experience were observed in four tests previously reported to induce fear-related responses in various animal species. A range of behavioural and physiological variables was recorded, such as latencies to approach the stimulus, time spent near stimulus and heart rate changes. Data were examined using correlation analysis and principal component analysis (PICA). Moderate to high correlations were found between some of the variables within and between the tests and a number of variables were identified in three components determined by the PCA. These three components accounted for 48% of the total variation and components 1 and 3 contained high loadings for the latency to approach and time spent near the stimulus in a novel object test and a startling test, respectively. Component 2 contained variables measuring latencies and entries to areas in a light/dark test and an elevated plus maze test. It is suggested that component 1 may be a measure of a response to novelty and component 3 may be a measure of a response to startling stimuli, while component 2 may be a measure of exploration. Components I and 3 may thus be appropriate measures of different aspects of fear. The suitability of each test and its limitations in measuring fear of novel and startling stimuli in the dog are discussed. (C) 2003 Elsevier Science B.V. All rights reserved.