This project aims to provide data on the contextual conditions involved in the processing of feared stimuli in individuals with specific phobias. Neuroimaging data are confirming the existence of two pathways in the processing of phobic stimuli. On the one hand, there is the short/non-conscious pathway, which involves a direct thalamus-amygdala connection; and on the other hand, there is the long/conscious pathway, with more complex thalamus-sensory cortex connections and the entorhinal cortex-hippocampus-subiculo-amygdala association. Both pathways produce the emotional response to phobic stimuli, but the existence of this second pathway may have clear implications for both the conceptualization of phobias and therapeutic processes. In this latter aspect, the long pathway, due to its more reflective nature, may enhance the effectiveness of exposure techniques to feared stimuli, resulting from improved deactivation techniques and greater efficacy of self-instruction training. In this sense, the aim is to understand what favors the emergence of the long pathway, in order to facilitate the effectiveness of the therapeutic approach. Specifically, this project will seek to determine, using phobias of small animals (cockroaches, spiders, lizards, and rats) as a criterion, whether the emergence of the conscious pathway is related to the nature of the stimuli (real images vs. virtual images) and/or their intensity (proximity of the stimulus and number of small animals); and whether it is related to contextual factors such as the use of self-instructions or deactivation processes (diaphragmatic breathing). Furthermore, it will also consider whether there might be a temperamental personality pattern that makes a person more vulnerable to activating or not activating this pathway. The design will consist of a 3x2x2 factorial design, with independent variables (IVs), two between-subjects and one within-subjects: type of therapeutic resource used (diaphragmatic breathing, adaptive self-instructions, and no resource); The nature of the presentation of the phobic stimuli (real videos versus videos of virtual images), and the intensity of the phobic stimuli (high and low, in relation to the proximity and number of small animals), will be a within-subjects independent variable. For functional magnetic resonance imaging (fMRI), the phobic stimuli will be presented randomly to all participants (depending on the phobia): one small phobic animal up close, one phobic animal at a distance, three phobic animals up close, and three phobic animals at a distance. The groups will consist of 20 people with specific phobias of small animals, resulting in a total sample of 120 participants. It is expected that a lower but sufficient stimulus intensity (low), less immersive stimulation (virtual stimuli), and the use of self-instructions and breathing exercises will favor the emergence of the long/conscious pathway. According to the emotional processing theory of fear, identifying these conditions could optimize the effectiveness of therapies based on exposure techniques.