Predictive representation learning on photoplethysmography (PPG) can breach causal information access because normalization, non‑local transforms, or companion views may rely on withheld samples, even when causal attention is used. We introduce PulseBound, a PPG representation learner that couples physiologically structured future‑beat prediction with an explicit stored‑window information boundary.\ \ A content‑independent cutoff separates the visible prefix from the prediction target. Prefix‑only normalization, suffix replacement before derived‑view construction, and aligned masking guarantee that encoder inputs depend solely on the visible prefix and the cutoff. This yields stored‑suffix invariance: with fixed model state, randomness, prefix and cutoff, altering the stored suffix cannot change the forecast context.\ \ A shared horizon‑conditioned head predicts nine rhythm and morphology descriptors for up to four extractor‑valid future beats, using elementwise validity masks; optional ECG‑derived pulse‑arrival‑time supervision is restricted to training.\ \ On MIMIC and VitalDB groups held out from PulseBound backbone pretraining, PulseBound reduces the nine‑state transformed‑space MAE relative to last‑visible‑beat persistence by 28.06% and 22.22% respectively, and improves MAE, MAE‑Skill, and Spearman correlation across all 40 source‑cutoff‑horizon cells. In a separate benchmark of seven models on 13 downstream tasks, PulseBound achieves the best mean on nine frozen linear‑probe and seven full‑fine‑tuning tasks.\ \ Stored‑suffix interventions cause zero recorded changes in forecast contexts or predictions, with zero suffix‑input gradients at audited precision, confirming the strictness of the information boundary.\ \ These findings separate three testable aspects of predictive physiological representation learning: information access, supervised future structure, and transferability.\ \ Review