Eye movements associated with recognition of affect in humans: implications for the detection of concealed firearm carrying

In prior work the effect of gun carrying on the mood (i.e., affective state) of the surveillance targets was assessed. In order to examine this effect, mock CCTV footage was generated of persons who acted as surveillance targets whilst concealing either a firearm or an innocuous object matched to the firearm (for a complete description of this aspect of the experiment see [1]). It was found that gun carrying was associated with increased dysphoria (i.e., an affective state characterized by hostility, anxiety or depression) which was in accordance with results from previous research [2]. In a consequent study [3] the participants (i.e., observers) watched this footage and estimated the emotional state of surveillance targets without knowledge about the presence of a firearm (i.e., Affect Detection Task, ADT). Likertscale questionnaires were used to provide the data about which cues convey information needed to identify the affective state in a firearm and nonfirearm carrier. In order to infer which parts of the body of a person carrying a concealed firearm are relevant for performing this task without relying on the observers’ consciously reported strategies only, as it was done in the previous study [3], here the observers performed the ADT task whilst their eye movements were recorded. The eyetracking technique is known to be able to clarify whether involuntary eyemovements are related to attention to emotionally arousing information presented in images [4], [5]. Applying this particular method in the current study is therefore believed to provide information about whether the perception of a particular affective state of carriers of concealed firearms is associated with a certain eyemovement pattern.