Respiratory sinus arrhythmia during biofeedback is linked to persistent improvements in attention, short-term memory, and positive self-referential episodic memory

FIGURE 1. Virtual reality biofeedback system. (A) During heart rate variability biofeedback training, which involved slow and rhythmic breathing, abdominal and thoracic circumference as well as an electrocardiogram were recorded through a respiration belt and surface electrodes (hidden under the shirt). Signals were processed by a computer and could be visualized in the virtual reality environment. (B) Participants in (1) the active control group followed the same training but received no biofeedback. Biofeedback was displayed by colors of the avatar (respiration) and sea (heart rate) that changed gradually from (2) green to (3) orange/red during exhalation or when the heart rate dropped, respectively, and vice versa. (C) The target of the training was, as exemplified here, to maintain synchrony between the respiration and heart rate time signals (i.e., similar color between the avatar and sea) and to increase the heart rate amplitude (i.e., stronger shades of green and red for the sea).

FIGURE 2. Course of study and procedure of the cognitive pre- and post-assessment. (A) Cognitive effects that were sustained 1 week after self-regulation training were assessed. (B) The experimental procedure was identical between groups and tests. For the self-reference episodic memory task parallel task versions were devised for the pre- and post-test, respectively.

FIGURE 3. Physiological measures. Physiological parameters were recorded directly before (baseline), during (training), and 5 min after (recovery) each training session. The small data points represent means of individual participants summarized over training sessions. The large data points reflect estimated marginal means summarized across participants and sessions that account for random subject effects and individual differences in the new Body Mass Index. Vertical lines indicate 95% CIs. The dashed line stands for the target breathing rate during training. Phases were compared between baseline and training as well as baseline and recovery for the BG (top lines) and CG (bottom lines), respectively. HRV, heart rate variability; RMSSD, root mean square of successive differences; BG, biofeedback group; CG, active control group; P2T-RSA, natural logarithm of the respiratory sinus arrhythmia calculated by the peak-to-trough method. ***p < 0.001. **p < 0.01. *p < 0.05.

FIGURE 4. Training effect on objective memory recollection. Objective recollection hit rate was assessed by the proportion of correct source recollection of studied words. EMMs and 95% CIs, derived from robust linear mixed effect models, are presented per group, reference condition, valence, and test. EMMs accounted for inter-participant differences in response reaction time at encoding and random subject effects. Pre-test values did not differ significantly (p > 0.14) between groups. Only recollections of positive items encoded with the self-reference condition (bold frame) revealed a significant group × test interaction effect. Training related improvements were found only in the BG. BG, biofeedback group; CG, active control group; EMM, estimated marginal means. ***p < 0.001.

FIGURE 5. The self-regulation-self-referential-memory (SR-SRM) model. Autonomic control can be exercised through HRV self-regulation training and thus affects pathways of the autonomic nervous system which reach the central nervous system. These pathways connect via the brainstem with several forebrain structures in the central autonomic network. Due to shared structures (most notably the vmPFC) further brain networks are influenced down the line which are involved in self-referential information processing (1.1). Likewise, affective processing is affected by functional changes related to limbic structures (1.2) which in turn influence self-referential processing (1.3). Firstly, self-referential encoding (2.1) provides a memory advantage to the encoded information (3.1). Secondly, self-referential processes are associated with the internal focus on thoughts that can trigger reactivation of a new memory or add novel associations to it, especially when the memory is related to the self (2.2), supporting its consolidation and retrieval (3.2). Consequently, strengthening self-referential processing via HRV self-regulation training may support voluntary retrieval of memory. In this context,…