‘She says, he says’: Does the sex of an instructor interact with the grammatical gender of targets in a perspective-taking task?

<jats:sec><jats:title>Aims and objectives:</jats:title><jats:p> It has been claimed that grammatical gender can influence the perception of objects as being potentially more ‘masculine’ or ‘feminine’. The present study investigated effects of facilitation or interference on object selection by speakers whose L1 marks grammatical gender even when selecting objects in an L2 (English) which does not mark grammatical gender. Additionally, and in order to establish whether bilingualism itself influenced performance owing to a proposed bilingual advantage in inhibitory control, we investigated whether bilinguals would be more efficient than monolinguals at taking the allocentric perspective and switching between perspectives. </jats:p></jats:sec><jats:sec><jats:title>Methodology:</jats:title><jats:p> Participants were asked to select objects by an instructor whose biological sex (and voice) was either congruent or incongruent with the grammatical gender of the object to be selected. Two groups of 16 bilinguals each were recruited on the basis of whether their L1s marked for grammatical gender or not, and a further group of 16 monolingual English speakers were tested as a control. </jats:p></jats:sec><jats:sec><jats:title>Data and analysis:</jats:title><jats:p> Groups were compared by means of mixed-design repeated measures ANOVAs with response times for target selections as the dependent variables. </jats:p></jats:sec><jats:sec><jats:title>Findings:</jats:title><jats:p> When tested in English, bilinguals whose L1 marked grammatical gender showed no effect of gender congruency in this task, nor did bilinguals outperform monolinguals in taking the allocentric perspective or in perspective switching. </jats:p></jats:sec><jats:sec><jats:title>Originality:</jats:title><jats:p> For the first time, potential grammatical gender effects were investigated on a task where the fast and accurate processing of real male and female voices is fundamental to the efficiency of object selection performance. </jats:p></jats:sec><jats:sec><jats:title>Implications:</jats:title><jats:p> The present findings are interpreted as evidence that the effects of L1 grammatical gender on tasks performed in an L2 do not extend to tasks where the link between biological sex and grammatical gender is not made explicit. </jats:p></jats:sec>