Cognitive neurosciences have made significant progress in learning about brain activity in situatedcognition, thanks to adopting stimuli that simulate immersion in naturalistic conditions insteadof isolated artificial stimuli. In particular, the use of films in neuroscientific experiments, aparadigm often referred to as neurocinematics, has contributed to this success. The use ofcinematic stimuli, however, has also revealed a fundamental shortcoming of neuroimaging studies:The lack of conceptual and methodological means to handle the viewers' experience of narrativeevents in their temporally extended contexts in the scale of full cinematic narrative, not tomention life itself. In order to give a conceptual structure to the issue of temporal contexts,we depart from the neurophenomenological approach to time consciousness by neurobiologistFrancisco Varela, which in turn builds on Husserl's phenomenology of time. More specifically, wewill discuss the experience of narrative tension, determined by backward-looking conceptualizingretention, and forward-looking anticipatory protention. Further, this conceptual structure isbuilt into a preliminary mathematical model, simulating the dynamics of decaying and refreshingmemory traces that aggregates a retentive perspective for each moment of nowness, which in turnmay trigger anticipations for coming events, in terms of Varela and Husserl, protentions. Thepresent tentative mathematical model is constructed using simple placeholder functions, with theintention that they would eventually be replaced by models based on empirical observations onthe psychological capabilities that support narrative sensemaking. The final goal is a model thatsuccessfully simulates the way how the memory system maintains narrative tension beyond thetransient nowness window, and thereby allows mappings to observed brain activity with a richtemporal system of narrative contexts.