Description
Quantifies how often and in what structure a behavioural time series revisits similar states, measuring determinism, stability and coordination.
A recurrence plot marks every pair of time points at which a system's state is similar. RQA quantifies the plot: recurrence rate, determinism (proportion of recurrent points forming diagonal lines, indicating repeated sequences), laminarity (vertical lines, indicating stuck states), entropy of line lengths and trapping time. Cross-recurrence quantification compares two time series, such as two collaborators or a user and an agent, to measure coordination and leading/following. RQA works on short, nonstationary, noisy data where classical methods fail, and has been applied in psychology to movement, eye gaze and conversation. For UX it offers measures of routine, rigidity and coordination in interaction streams.
Attributes
- Discipline
- Information Theory, Physics & Mathematics
- Loss type
- Unmigrated
- Goal
- Measure behaviour
- Stage
- Post-launch
- Data
- Quantitative
- Effort
- Weeks
Steps
- Choose the behavioural time series (e.g. action categories, cursor positions, gaze, speech features).
- For continuous data, choose embedding dimension and delay (false nearest neighbours, mutual information); for categorical data use direct matching.
- Set the radius threshold to achieve a target recurrence rate (e.g. 2-5%).
- Compute RQA measures: recurrence rate, determinism, laminarity, trapping time, entropy.
- Compare measures across conditions, users or time, with surrogate data tests to check against randomness.
- Use cross-recurrence for dyads (e.g. human-AI turn-taking) to measure coordination and lag.
Sources
- Marwan et al., Recurrence plots for the analysis of complex systems (2007), Physics Reports
- Webber & Zbilut, Recurrence quantification analysis of nonlinear dynamical systems (2005), in Riley & Van Orden (eds.), Tutorials in Contemporary Nonlinear Methods for the Behavioral Sciences
Also known as
- RQA
- Recurrence plots
- Cross-recurrence quantification
Categories
- Information Theory, Physics & Mathematics