The scientific debate about the possibility to identify the specific psychophysiological patterns correlated with distinct emotions is still open. In this preliminary research work, we explored cardiac and respiration dynamics while feeling the following affective states: sadness, happiness, anger and fear. Electrocardiogram (ECG) and respiration signals were recorded from 27 normal students who voluntarily participated in the experiment, where they had to recall and relive an important autobiographical episode for each target emotion after a beginning session of baseline. Preliminary statistical analyses about cardiac and respiration indexes show that the four emotions are significantly different from the baseline conditions, and that certain measures can distinguish between the different emotions too. These results have to be taken as a partial anticipation of the ongoing research, since the study is still a work in progress with the aim to increase the number of subjects representing the experimental sample up to 60 participants. Nevertheless, this study shows how the modifications in terms of somatic activity might become a source of information for the definition of specific biological patterns while feeling the target emotions, allowing to explore the application of automatic emotion recognition in applied fields such as communication