Human handling and presentation of a novel object evoke independent dimensions of fear in Japanese quail

Abstract

Fear is a concept comprising several dimensions, but the nature of these dimensions and the relationships between them remain elusive. To investigate these dimensions in birds, we have used two genetic lines of quail divergently selected on tonic immobility duration, a behavioural index of fear. These two lines differ in their behavioural response to some, but not all, fear-inducing situations. In the present study, we investigated the contribution of human intervention in the differentiation between the two lines. To do this, fear responses towards a novel object were compared between lines in three conditions: (1) in the home cage without any human intervention, (2) in the home cage after human handling and (3) after placement in a novel environment by human handling. Fear behaviour differed between lines after human handling, with or without placement in a novel environment, but presentation of a novel object in the home cage without any human intervention induced similar fear responses in the two lines of quail. These results lead us to suggest that in quail, human intervention evokes a dimension of fear that differs from that evoked by sudden presentation of a novel object, in that these two dimensions may be selected independently.

Introduction

Genetic selection on fear behaviour has proved useful not only to obtain animal models helping to understand the mechanisms of fear behaviour, but also to reduce the expression of undesirable fear behaviour in farm species (Landgraf and Wigger, 2002, Faure and Mills, 1998). The success of such experiments has strengthened the initial assumption that selection on a single behavioural fear trait affected fearfulness in general, i.e. the propensity to be more or less easily frightened (Faure et al., 2006, Broadhurst, 1975, Landgraf and Wigger, 2002, Jones, 1996). In this regard, fear has traditionally been considered as a unitary concept. However, this view has recently been challenged by data in birds and mammals, which have uncovered the complexity of this seemingly simple concept (see for example Russell, 2003, Campler et al., 2009, López-Aumatell et al., 2009). In birds, evidence notably came from two lines of Japanese quail that have been divergently selected for over 40 generations on tonic immobility duration, a behavioural index of fear, and that are considered to differ in their inherent fearfulness (Mills and Faure, 1991, Faure et al., 2006). In a number of classical behavioural tests, including the ‘open-field’, the ‘capture’ and the ‘hole-in-the-wall’ box tests, quail of the long tonic immobility line (LTI) exhibit consistently higher levels of fear than quail of the short tonic immobility line (STI) (Faure et al., 2006). However, quail of both lines have been shown to display similar fear reactions when presented with a novel object or a novel sound in their home cage (Valance et al., 2007, Richard et al., 2008, Saint-Dizier et al., 2008), suggesting that novelty-induced fear responses have not been affected by selection on tonic immobility. Thus it appears that genetic selection on a behavioural fear trait does not necessarily affect all dimensions of fear.

Although the multidimensional nature of fear has been recognised by authors from various backgrounds, there is no consensus about what the underlying dimensions may actually be in practice (Russell, 2003). In the present study, we used the term ‘dimensions’ to name distinct emotions that share the definition of fear as an emotion triggered by the perception of a danger (Jones, 1996). Thus, we considered that the concept of fear comprised several emotions and that distinct dimensions of fear might be evoked independently and should differ in their genetic substrate. In an attempt to clarify these concepts, we sought to experimentally characterise the dimensions of fear that were affected by genetic selection in the STI and LTI lines of quail. We hypothesised that the choice of the criterion used for selecting the individuals, the duration of tonic immobility, may have played a critical role in determining the outcoming phenotype. Tonic immobility is an unlearned fear response characterised by profound motor inhibition (Mills and Faure, 1991) and it is experimentally induced by physical restraint in a soundproof room. Thus, two major characteristics of the selection test are placement in a novel environment and obvious human intervention. Both of these factors, or at least one of them (human intervention), were present in the fear tests in which LTI quail have been shown to display higher levels of fear than STI quail, notably the open-field and ‘hole-in-the-wall’ box tests as well as restraint in a ‘crush cage’ and capture by a human (Jones et al., 1991, Jones et al., 1994, Mills and Faure, 2000). In contrast, in the home cage and in the absence of human interference, presentation of a novel object or of a novel sound induced similar behavioural fear reactions in the two lines of quail (Valance et al., 2007, Saint-Dizier et al., 2008, Richard et al., 2008). These observations have led us to suggest that human intervention, combined or not with exposure to a novel environment, might be a critical factor in revealing the differences in behaviour between STI and LTI quail. In other words, we hypothesised that selection on tonic immobility duration affected a dimension of fear evoked by human intervention, but not the dimension evoked by a novel object.

In the present experiments we sought to investigate the contribution of human handling (in the form of a single, brief manipulation of the quail by the experimenter) and placement in a novel environment in the behavioural differentiation between STI and LTI quail. To do this, two experiments with quail from the same hatches were carried out in parallel. The quail were subjected to three variants of the novel object test: in the home cage without any human intervention in the cage, in the home cage after human handling (Experiment 1) and after placement in a novel environment by human handling (Experiment 2). We hypothesised that the behavioural responses of STI and LTI quail induced by presentation of a novel object after human handling and placement in a novel environment might differ, whereas STI and LTI quail should present similar fear responses when the novel object is presented in the home cage without any human intervention.