A Continuous Approach to Coding Expressed Affect During Parent-Child Conflict and Repair

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Parent-child conflict and repair are important developmental processes associated with later relationship quality and mental health outcomes (Li et al., 2026). Examining how emotion is communicated during conflict and repair is critical to understanding how families navigate these processes. Real-time behavioural coding of emotional expressions allows researchers to investigate this; however, behavioural coding systems used to capture affect have approached coding emotions as discrete phenomena, creating three limitations: (1) differences in expressed intensity are lost, such that mild annoyance and yelling may both be coded as anger; (2) categorical boundaries can become ambiguous when multiple affective states are expressed simultaneously; and (3) discrete variables limit the statistical approaches available for examining affective dynamics. My project aimed to address this gap by formalising the Continuous Rating of Affective Valence of Expression (CRAVE) coding system, a preliminary approach developed in the Adolescent Dynamics Lab for capturing affect as a continuous variable (Martin et al., 2026). The formalised CRAVE code captures nonverbal affective expression using the gestalt of tone of voice, body language, and facial expression. Scores range from -10 to +10, with the sign indicating valence and larger absolute values indicating greater intensity. Two undergraduate coders applied the revised manual to conflict and repair discussions from 95 parent-adolescent dyads, recording ratings via joystick using the CARMA coding platform (Girard, 2014). Interrater reliability was assessed using intraclass correlation coefficients (ICC) for consistency and mean triangular rWG, with thresholds of .75 and .95, respectively. We achieved high interrater reliability, with a mean ICC of .86 and a mean triangular rWG of .99. My project supports the reliability of CRAVE as a tool for examining continuous affective dynamics during parent-adolescent conflict and repair. Beginning this fall, CRAVE will be compared with established discrete emotion coding systems to examine the added value of continuous affective measurement.Faculty Supervisor: Dr. Tom Hollenstein References Girard, J. M. (2014). CARMA: Software for continuous affect rating and media annotation. Journal of Open Research Software, 2(1), e5. https://doi.org/10.5334/jors.ar Li, X., Lougheed, J. P., & Hollenstein, T. (2026). From friction to flow: Dyadic affective flexibility during and after conflicts predicts trajectories of mother–adolescent relationships. Developmental Psychology, 62(5), 915–931. https://doi.org/10.1037/dev0001939 Martin, V. E., Lougheed, J. P., & Hollenstein, T. (2026). Real‐Time Contingencies of Emotion Socialization During Parent–Adolescent Conflicts. Social Development (Oxford, England), 35(1). https://doi.org/10.1111/sode.70045