Aims: Emerging evidence indicates the existence of bidirectional, complex, and nonlinear communication between heartbeat and brain dynamics, which provides support for long-standing hypotheses regarding the causal relationship between physiological activity and emotions. Therefore, this study aimed to leverage network physiology and brain-heart interaction (BHI) principles as a way to represent complex emotional behavior, paving the way for explainable affective computing. Method: The controlled time delay stability algorithm (cTDS) is used to investigate correlations between photoplethysmography (PPG) features and electroencephalography (EEG) signals for distinguishing between four affective states based on High (H) and low (L) valence (V) and arousal (A). PPG parameters (amplitude, peak-to-peak interval (PPI), and pulse width amplitude (PWA)) and EEG data from eight electrodes (fp1, fp2, c3, c4, o1, o2, t7, t8) across four frequency bands (alpha, beta, gamma, theta) were analyzed using the DEAP dataset. Results: Significant correlations were found between EEG frequency bands and both PWA and PPI across affective states (p < 0.01). However, reverse relationships, where PPI or PWA influenced EEG bands, did not show significance across the affective states. Additionally, the correlation of PPG amplitude and EEG bands, or vice versa, did not differentiate between affective states significantly, indicating that PPG amplitude may not be as indicative of affective states as PPI or PWA in this context. Conclusion: The findings highlight EEG-PWA and EEG-PPI connectivity as reliable indicators of affective states. Unlike PA, which reflects systolic blood pressure changes, PPI and PWA offer unique insights into cardiac autonomic activity and vascular tone regulation, closely linked to emotional processing. The inability of EEG-PA connectivity to differentiate between affective states suggests that changes in systolic blood pressure may not be intricately tied to the nuanced variations in affective states captured in this study. Leveraging PWA and PPI promises to refine affective computing and human-computer interaction.